- 
-
-
- 
- 
- 
-
-
List of implemented papers
- -* **Convolutional Neural Networks (CNN)** - + [Language Modeling with Gated Convolutional Networks (Dauphin et al., 2017)](examples/language_model/conv_lm/README.md) - + [Convolutional Sequence to Sequence Learning (Gehring et al., 2017)](examples/conv_seq2seq/README.md) - + [Classical Structured Prediction Losses for Sequence to Sequence Learning (Edunov et al., 2018)](https://github.com/pytorch/fairseq/tree/classic_seqlevel) - + [Hierarchical Neural Story Generation (Fan et al., 2018)](examples/stories/README.md) - + [wav2vec: Unsupervised Pre-training for Speech Recognition (Schneider et al., 2019)](examples/wav2vec/README.md) -* **LightConv and DynamicConv models** - + [Pay Less Attention with Lightweight and Dynamic Convolutions (Wu et al., 2019)](examples/pay_less_attention_paper/README.md) -* **Long Short-Term Memory (LSTM) networks** - + Effective Approaches to Attention-based Neural Machine Translation (Luong et al., 2015) -* **Transformer (self-attention) networks** - + Attention Is All You Need (Vaswani et al., 2017) - + [Scaling Neural Machine Translation (Ott et al., 2018)](examples/scaling_nmt/README.md) - + [Understanding Back-Translation at Scale (Edunov et al., 2018)](examples/backtranslation/README.md) - + [Adaptive Input Representations for Neural Language Modeling (Baevski and Auli, 2018)](examples/language_model/README.adaptive_inputs.md) - + [Lexically constrained decoding with dynamic beam allocation (Post & Vilar, 2018)](examples/constrained_decoding/README.md) - + [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context (Dai et al., 2019)](examples/truncated_bptt/README.md) - + [Adaptive Attention Span in Transformers (Sukhbaatar et al., 2019)](examples/adaptive_span/README.md) - + [Mixture Models for Diverse Machine Translation: Tricks of the Trade (Shen et al., 2019)](examples/translation_moe/README.md) - + [RoBERTa: A Robustly Optimized BERT Pretraining Approach (Liu et al., 2019)](examples/roberta/README.md) - + [Facebook FAIR's WMT19 News Translation Task Submission (Ng et al., 2019)](examples/wmt19/README.md) - + [Jointly Learning to Align and Translate with Transformer Models (Garg et al., 2019)](examples/joint_alignment_translation/README.md ) - + [Multilingual Denoising Pre-training for Neural Machine Translation (Liu et at., 2020)](examples/mbart/README.md) - + [Neural Machine Translation with Byte-Level Subwords (Wang et al., 2020)](examples/byte_level_bpe/README.md) - + [Unsupervised Quality Estimation for Neural Machine Translation (Fomicheva et al., 2020)](examples/unsupervised_quality_estimation/README.md) - + [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations (Baevski et al., 2020)](examples/wav2vec/README.md) - + [Generating Medical Reports from Patient-Doctor Conversations Using Sequence-to-Sequence Models (Enarvi et al., 2020)](examples/pointer_generator/README.md) - + [Linformer: Self-Attention with Linear Complexity (Wang et al., 2020)](examples/linformer/README.md) - + [Cross-lingual Retrieval for Iterative Self-Supervised Training (Tran et al., 2020)](examples/criss/README.md) - + [Deep Transformers with Latent Depth (Li et al., 2020)](examples/latent_depth/README.md) - + [Unsupervised Cross-lingual Representation Learning for Speech Recognition (Conneau et al., 2020)](https://arxiv.org/abs/2006.13979) - + [Robust wav2vec 2.0: Analyzing Domain Shift in Self-Supervised Pre-Training (Hsu, et al., 2021)](https://arxiv.org/abs/2104.01027) - + [Unsupervised Speech Recognition (Baevski, et al., 2021)](https://arxiv.org/abs/2105.11084) -* **Non-autoregressive Transformers** - + Non-Autoregressive Neural Machine Translation (Gu et al., 2017) - + Deterministic Non-Autoregressive Neural Sequence Modeling by Iterative Refinement (Lee et al. 2018) - + Insertion Transformer: Flexible Sequence Generation via Insertion Operations (Stern et al. 2019) - + Mask-Predict: Parallel Decoding of Conditional Masked Language Models (Ghazvininejad et al., 2019) - + [Levenshtein Transformer (Gu et al., 2019)](examples/nonautoregressive_translation/README.md) -* **Finetuning** - + [Better Fine-Tuning by Reducing Representational Collapse (Aghajanyan et al. 2020)](examples/rxf/README.md) - -
Previous updates
- -* September 2020: [Added Linformer code](examples/linformer/README.md) -* September 2020: [Added pointer-generator networks](examples/pointer_generator/README.md) -* August 2020: [Added lexically constrained decoding](examples/constrained_decoding/README.md) -* August 2020: [wav2vec2 models and code released](examples/wav2vec/README.md) -* July 2020: [Unsupervised Quality Estimation code released](examples/unsupervised_quality_estimation/README.md) -* May 2020: [Follow fairseq on Twitter](https://twitter.com/fairseq) -* April 2020: [Monotonic Multihead Attention code released](examples/simultaneous_translation/README.md) -* April 2020: [Quant-Noise code released](examples/quant_noise/README.md) -* April 2020: [Initial model parallel support and 11B parameters unidirectional LM released](examples/megatron_11b/README.md) -* March 2020: [Byte-level BPE code released](examples/byte_level_bpe/README.md) -* February 2020: [mBART model and code released](examples/mbart/README.md) -* February 2020: [Added tutorial for back-translation](https://github.com/pytorch/fairseq/tree/main/examples/backtranslation#training-your-own-model-wmt18-english-german) -* December 2019: [fairseq 0.9.0 released](https://github.com/pytorch/fairseq/releases/tag/v0.9.0) -* November 2019: [VizSeq released (a visual analysis toolkit for evaluating fairseq models)](https://facebookresearch.github.io/vizseq/docs/getting_started/fairseq_example) -* November 2019: [CamemBERT model and code released](examples/camembert/README.md) -* November 2019: [BART model and code released](examples/bart/README.md) -* November 2019: [XLM-R models and code released](examples/xlmr/README.md) -* September 2019: [Nonautoregressive translation code released](examples/nonautoregressive_translation/README.md) -* August 2019: [WMT'19 models released](examples/wmt19/README.md) -* July 2019: fairseq relicensed under MIT license -* July 2019: [RoBERTa models and code released](examples/roberta/README.md) -* June 2019: [wav2vec models and code released](examples/wav2vec/README.md) - -
([Edunov et al., 2018](https://arxiv.org/abs/1808.09381))
WMT'18 winner | [WMT'18 English-German](http://www.statmt.org/wmt18/translation-task.html) | [download (.tar.gz)](https://dl.fbaipublicfiles.com/fairseq/models/wmt18.en-de.ensemble.tar.gz)
See NOTE in the archive - -## Example usage (torch.hub) - -We require a few additional Python dependencies for preprocessing: -```bash -pip install subword_nmt sacremoses -``` - -Then to generate translations from the full model ensemble: -```python -import torch - -# List available models -torch.hub.list('pytorch/fairseq') # [..., 'transformer.wmt18.en-de', ... ] - -# Load the WMT'18 En-De ensemble -en2de_ensemble = torch.hub.load( - 'pytorch/fairseq', 'transformer.wmt18.en-de', - checkpoint_file='wmt18.model1.pt:wmt18.model2.pt:wmt18.model3.pt:wmt18.model4.pt:wmt18.model5.pt', - tokenizer='moses', bpe='subword_nmt') - -# The ensemble contains 5 models -len(en2de_ensemble.models) -# 5 - -# Translate -en2de_ensemble.translate('Hello world!') -# 'Hallo Welt!' -``` - -## Training your own model (WMT'18 English-German) - -The following instructions can be adapted to reproduce the models from the paper. - - -#### Step 1. Prepare parallel data and optionally train a baseline (English-German) model - -First download and preprocess the data: -```bash -# Download and prepare the data -cd examples/backtranslation/ -bash prepare-wmt18en2de.sh -cd ../.. - -# Binarize the data -TEXT=examples/backtranslation/wmt18_en_de -fairseq-preprocess \ - --joined-dictionary \ - --source-lang en --target-lang de \ - --trainpref $TEXT/train --validpref $TEXT/valid --testpref $TEXT/test \ - --destdir data-bin/wmt18_en_de --thresholdtgt 0 --thresholdsrc 0 \ - --workers 20 - -# Copy the BPE code into the data-bin directory for future use -cp examples/backtranslation/wmt18_en_de/code data-bin/wmt18_en_de/code -``` - -(Optionally) Train a baseline model (English-German) using just the parallel data: -```bash -CHECKPOINT_DIR=checkpoints_en_de_parallel -fairseq-train --fp16 \ - data-bin/wmt18_en_de \ - --source-lang en --target-lang de \ - --arch transformer_wmt_en_de_big --share-all-embeddings \ - --dropout 0.3 --weight-decay 0.0 \ - --criterion label_smoothed_cross_entropy --label-smoothing 0.1 \ - --optimizer adam --adam-betas '(0.9, 0.98)' --clip-norm 0.0 \ - --lr 0.001 --lr-scheduler inverse_sqrt --warmup-updates 4000 \ - --max-tokens 3584 --update-freq 16 \ - --max-update 30000 \ - --save-dir $CHECKPOINT_DIR -# Note: the above command assumes 8 GPUs. Adjust `--update-freq` if you have a -# different number of GPUs. -``` - -Average the last 10 checkpoints: -```bash -python scripts/average_checkpoints.py \ - --inputs $CHECKPOINT_DIR \ - --num-epoch-checkpoints 10 \ - --output $CHECKPOINT_DIR/checkpoint.avg10.pt -``` - -Evaluate BLEU: -```bash -# tokenized BLEU on newstest2017: -bash examples/backtranslation/tokenized_bleu.sh \ - wmt17 \ - en-de \ - data-bin/wmt18_en_de \ - data-bin/wmt18_en_de/code \ - $CHECKPOINT_DIR/checkpoint.avg10.pt -# BLEU4 = 29.57, 60.9/35.4/22.9/15.5 (BP=1.000, ratio=1.014, syslen=63049, reflen=62152) -# compare to 29.46 in Table 1, which is also for tokenized BLEU - -# generally it's better to report (detokenized) sacrebleu though: -bash examples/backtranslation/sacrebleu.sh \ - wmt17 \ - en-de \ - data-bin/wmt18_en_de \ - data-bin/wmt18_en_de/code \ - $CHECKPOINT_DIR/checkpoint.avg10.pt -# BLEU+case.mixed+lang.en-de+numrefs.1+smooth.exp+test.wmt17+tok.13a+version.1.4.3 = 29.0 60.6/34.7/22.4/14.9 (BP = 1.000 ratio = 1.013 hyp_len = 62099 ref_len = 61287) -``` - - -#### Step 2. Back-translate monolingual German data - -Train a reverse model (German-English) to do the back-translation: -```bash -CHECKPOINT_DIR=checkpoints_de_en_parallel -fairseq-train --fp16 \ - data-bin/wmt18_en_de \ - --source-lang de --target-lang en \ - --arch transformer_wmt_en_de_big --share-all-embeddings \ - --dropout 0.3 --weight-decay 0.0 \ - --criterion label_smoothed_cross_entropy --label-smoothing 0.1 \ - --optimizer adam --adam-betas '(0.9, 0.98)' --clip-norm 0.0 \ - --lr 0.001 --lr-scheduler inverse_sqrt --warmup-updates 4000 \ - --max-tokens 3584 --update-freq 16 \ - --max-update 30000 \ - --save-dir $CHECKPOINT_DIR -# Note: the above command assumes 8 GPUs. Adjust `--update-freq` if you have a -# different number of GPUs. -``` - -Let's evaluate the back-translation (BT) model to make sure it is well trained: -```bash -bash examples/backtranslation/sacrebleu.sh \ - wmt17 \ - de-en \ - data-bin/wmt18_en_de \ - data-bin/wmt18_en_de/code \ - $CHECKPOINT_DIR/checkpoint_best.py -# BLEU+case.mixed+lang.de-en+numrefs.1+smooth.exp+test.wmt17+tok.13a+version.1.4.3 = 34.9 66.9/41.8/28.5/19.9 (BP = 0.983 ratio = 0.984 hyp_len = 63342 ref_len = 64399) -# compare to the best system from WMT'17 which scored 35.1: http://matrix.statmt.org/matrix/systems_list/1868 -``` - -Next prepare the monolingual data: -```bash -# Download and prepare the monolingual data -# By default the script samples 25M monolingual sentences, which after -# deduplication should be just over 24M sentences. These are split into 25 -# shards, each with 1M sentences (except for the last shard). -cd examples/backtranslation/ -bash prepare-de-monolingual.sh -cd ../.. - -# Binarize each shard of the monolingual data -TEXT=examples/backtranslation/wmt18_de_mono -for SHARD in $(seq -f "%02g" 0 24); do \ - fairseq-preprocess \ - --only-source \ - --source-lang de --target-lang en \ - --joined-dictionary \ - --srcdict data-bin/wmt18_en_de/dict.de.txt \ - --testpref $TEXT/bpe.monolingual.dedup.${SHARD} \ - --destdir data-bin/wmt18_de_mono/shard${SHARD} \ - --workers 20; \ - cp data-bin/wmt18_en_de/dict.en.txt data-bin/wmt18_de_mono/shard${SHARD}/; \ -done -``` - -Now we're ready to perform back-translation over the monolingual data. The -following command generates via sampling, but it's possible to use greedy -decoding (`--beam 1`), beam search (`--beam 5`), -top-k sampling (`--sampling --beam 1 --sampling-topk 10`), etc.: -```bash -mkdir backtranslation_output -for SHARD in $(seq -f "%02g" 0 24); do \ - fairseq-generate --fp16 \ - data-bin/wmt18_de_mono/shard${SHARD} \ - --path $CHECKPOINT_DIR/checkpoint_best.pt \ - --skip-invalid-size-inputs-valid-test \ - --max-tokens 4096 \ - --sampling --beam 1 \ - > backtranslation_output/sampling.shard${SHARD}.out; \ -done -``` - -After BT, use the `extract_bt_data.py` script to re-combine the shards, extract -the back-translations and apply length ratio filters: -```bash -python examples/backtranslation/extract_bt_data.py \ - --minlen 1 --maxlen 250 --ratio 1.5 \ - --output backtranslation_output/bt_data --srclang en --tgtlang de \ - backtranslation_output/sampling.shard*.out - -# Ensure lengths are the same: -# wc -l backtranslation_output/bt_data.{en,de} -# 21795614 backtranslation_output/bt_data.en -# 21795614 backtranslation_output/bt_data.de -# 43591228 total -``` - -Binarize the filtered BT data and combine it with the parallel data: -```bash -TEXT=backtranslation_output -fairseq-preprocess \ - --source-lang en --target-lang de \ - --joined-dictionary \ - --srcdict data-bin/wmt18_en_de/dict.en.txt \ - --trainpref $TEXT/bt_data \ - --destdir data-bin/wmt18_en_de_bt \ - --workers 20 - -# We want to train on the combined data, so we'll symlink the parallel + BT data -# in the wmt18_en_de_para_plus_bt directory. We link the parallel data as "train" -# and the BT data as "train1", so that fairseq will combine them automatically -# and so that we can use the `--upsample-primary` option to upsample the -# parallel data (if desired). -PARA_DATA=$(readlink -f data-bin/wmt18_en_de) -BT_DATA=$(readlink -f data-bin/wmt18_en_de_bt) -COMB_DATA=data-bin/wmt18_en_de_para_plus_bt -mkdir -p $COMB_DATA -for LANG in en de; do \ - ln -s ${PARA_DATA}/dict.$LANG.txt ${COMB_DATA}/dict.$LANG.txt; \ - for EXT in bin idx; do \ - ln -s ${PARA_DATA}/train.en-de.$LANG.$EXT ${COMB_DATA}/train.en-de.$LANG.$EXT; \ - ln -s ${BT_DATA}/train.en-de.$LANG.$EXT ${COMB_DATA}/train1.en-de.$LANG.$EXT; \ - ln -s ${PARA_DATA}/valid.en-de.$LANG.$EXT ${COMB_DATA}/valid.en-de.$LANG.$EXT; \ - ln -s ${PARA_DATA}/test.en-de.$LANG.$EXT ${COMB_DATA}/test.en-de.$LANG.$EXT; \ - done; \ -done -``` - - -#### 3. Train an English-German model over the combined parallel + BT data - -Finally we can train a model over the parallel + BT data: -```bash -CHECKPOINT_DIR=checkpoints_en_de_parallel_plus_bt -fairseq-train --fp16 \ - data-bin/wmt18_en_de_para_plus_bt \ - --upsample-primary 16 \ - --source-lang en --target-lang de \ - --arch transformer_wmt_en_de_big --share-all-embeddings \ - --dropout 0.3 --weight-decay 0.0 \ - --criterion label_smoothed_cross_entropy --label-smoothing 0.1 \ - --optimizer adam --adam-betas '(0.9, 0.98)' --clip-norm 0.0 \ - --lr 0.0007 --lr-scheduler inverse_sqrt --warmup-updates 4000 \ - --max-tokens 3584 --update-freq 16 \ - --max-update 100000 \ - --save-dir $CHECKPOINT_DIR -# Note: the above command assumes 8 GPUs. Adjust `--update-freq` if you have a -# different number of GPUs. -``` - -Average the last 10 checkpoints: -```bash -python scripts/average_checkpoints.py \ - --inputs $CHECKPOINT_DIR \ - --num-epoch-checkpoints 10 \ - --output $CHECKPOINT_DIR/checkpoint.avg10.pt -``` - -Evaluate BLEU: -```bash -# tokenized BLEU on newstest2017: -bash examples/backtranslation/tokenized_bleu.sh \ - wmt17 \ - en-de \ - data-bin/wmt18_en_de \ - data-bin/wmt18_en_de/code \ - $CHECKPOINT_DIR/checkpoint.avg10.pt -# BLEU4 = 32.35, 64.4/38.9/26.2/18.3 (BP=0.977, ratio=0.977, syslen=60729, reflen=62152) -# compare to 32.35 in Table 1, which is also for tokenized BLEU - -# generally it's better to report (detokenized) sacrebleu: -bash examples/backtranslation/sacrebleu.sh \ - wmt17 \ - en-de \ - data-bin/wmt18_en_de \ - data-bin/wmt18_en_de/code \ - $CHECKPOINT_DIR/checkpoint.avg10.pt -# BLEU+case.mixed+lang.en-de+numrefs.1+smooth.exp+test.wmt17+tok.13a+version.1.4.3 = 31.5 64.3/38.2/25.6/17.6 (BP = 0.971 ratio = 0.971 hyp_len = 59515 ref_len = 61287) -``` - - -## Citation -```bibtex -@inproceedings{edunov2018backtranslation, - title = {Understanding Back-Translation at Scale}, - author = {Edunov, Sergey and Ott, Myle and Auli, Michael and Grangier, David}, - booktitle = {Conference of the Association for Computational Linguistics (ACL)}, - year = 2018, -} -``` diff --git a/fairseq/examples/backtranslation/deduplicate_lines.py b/fairseq/examples/backtranslation/deduplicate_lines.py deleted file mode 100644 index 50e458328c80b71c42a66d473381ca7e98d294da..0000000000000000000000000000000000000000 --- a/fairseq/examples/backtranslation/deduplicate_lines.py +++ /dev/null @@ -1,41 +0,0 @@ -#!/usr/bin/python3 -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. - -import argparse -import fileinput -import hashlib -import sys -from multiprocessing import Pool - - -def get_hashes_and_lines(raw_line): - hash = hashlib.md5(raw_line).hexdigest() - return hash, raw_line - - -def main(): - parser = argparse.ArgumentParser() - parser.add_argument("--workers", type=int, default=10) - parser.add_argument("files", nargs="*", help="input files") - args = parser.parse_args() - - seen = set() - with fileinput.input(args.files, mode="rb") as h: - pool = Pool(args.workers) - results = pool.imap_unordered(get_hashes_and_lines, h, 1000) - for i, (hash, raw_line) in enumerate(results): - if hash not in seen: - seen.add(hash) - sys.stdout.buffer.write(raw_line) - if i % 1000000 == 0: - print(i, file=sys.stderr, end="", flush=True) - elif i % 100000 == 0: - print(".", file=sys.stderr, end="", flush=True) - print(file=sys.stderr, flush=True) - - -if __name__ == "__main__": - main() diff --git a/fairseq/examples/backtranslation/extract_bt_data.py b/fairseq/examples/backtranslation/extract_bt_data.py deleted file mode 100644 index e766391e873d0d9a9561d67d5864934b2fad0681..0000000000000000000000000000000000000000 --- a/fairseq/examples/backtranslation/extract_bt_data.py +++ /dev/null @@ -1,72 +0,0 @@ -#!/usr/bin/env python -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. - -import argparse -import fileinput - -from tqdm import tqdm - - -def main(): - parser = argparse.ArgumentParser( - description=( - "Extract back-translations from the stdout of fairseq-generate. " - "If there are multiply hypotheses for a source, we only keep the first one. " - ) - ) - parser.add_argument("--output", required=True, help="output prefix") - parser.add_argument( - "--srclang", required=True, help="source language (extracted from H-* lines)" - ) - parser.add_argument( - "--tgtlang", required=True, help="target language (extracted from S-* lines)" - ) - parser.add_argument("--minlen", type=int, help="min length filter") - parser.add_argument("--maxlen", type=int, help="max length filter") - parser.add_argument("--ratio", type=float, help="ratio filter") - parser.add_argument("files", nargs="*", help="input files") - args = parser.parse_args() - - def validate(src, tgt): - srclen = len(src.split(" ")) if src != "" else 0 - tgtlen = len(tgt.split(" ")) if tgt != "" else 0 - if ( - (args.minlen is not None and (srclen < args.minlen or tgtlen < args.minlen)) - or ( - args.maxlen is not None - and (srclen > args.maxlen or tgtlen > args.maxlen) - ) - or ( - args.ratio is not None - and (max(srclen, tgtlen) / float(min(srclen, tgtlen)) > args.ratio) - ) - ): - return False - return True - - def safe_index(toks, index, default): - try: - return toks[index] - except IndexError: - return default - - with open(args.output + "." + args.srclang, "w") as src_h, open( - args.output + "." + args.tgtlang, "w" - ) as tgt_h: - for line in tqdm(fileinput.input(args.files)): - if line.startswith("S-"): - tgt = safe_index(line.rstrip().split("\t"), 1, "") - elif line.startswith("H-"): - if tgt is not None: - src = safe_index(line.rstrip().split("\t"), 2, "") - if validate(src, tgt): - print(src, file=src_h) - print(tgt, file=tgt_h) - tgt = None - - -if __name__ == "__main__": - main() diff --git a/fairseq/examples/backtranslation/prepare-de-monolingual.sh b/fairseq/examples/backtranslation/prepare-de-monolingual.sh deleted file mode 100644 index 5e67b2b3bcf27d3436031453e796e58a0ae79ec4..0000000000000000000000000000000000000000 --- a/fairseq/examples/backtranslation/prepare-de-monolingual.sh +++ /dev/null @@ -1,98 +0,0 @@ -#!/bin/bash - -SCRIPTS=mosesdecoder/scripts -TOKENIZER=$SCRIPTS/tokenizer/tokenizer.perl -NORM_PUNC=$SCRIPTS/tokenizer/normalize-punctuation.perl -REM_NON_PRINT_CHAR=$SCRIPTS/tokenizer/remove-non-printing-char.perl -BPEROOT=subword-nmt/subword_nmt - - -BPE_CODE=wmt18_en_de/code -SUBSAMPLE_SIZE=25000000 -LANG=de - - -OUTDIR=wmt18_${LANG}_mono -orig=orig -tmp=$OUTDIR/tmp -mkdir -p $OUTDIR $tmp - - -URLS=( - "http://www.statmt.org/wmt14/training-monolingual-news-crawl/news.2007.de.shuffled.gz" - "http://www.statmt.org/wmt14/training-monolingual-news-crawl/news.2008.de.shuffled.gz" - "http://www.statmt.org/wmt14/training-monolingual-news-crawl/news.2009.de.shuffled.gz" - "http://www.statmt.org/wmt14/training-monolingual-news-crawl/news.2010.de.shuffled.gz" - "http://www.statmt.org/wmt14/training-monolingual-news-crawl/news.2011.de.shuffled.gz" - "http://www.statmt.org/wmt14/training-monolingual-news-crawl/news.2012.de.shuffled.gz" - "http://www.statmt.org/wmt14/training-monolingual-news-crawl/news.2013.de.shuffled.gz" - "http://www.statmt.org/wmt15/training-monolingual-news-crawl-v2/news.2014.de.shuffled.v2.gz" - "http://data.statmt.org/wmt16/translation-task/news.2015.de.shuffled.gz" - "http://data.statmt.org/wmt17/translation-task/news.2016.de.shuffled.gz" - "http://data.statmt.org/wmt18/translation-task/news.2017.de.shuffled.deduped.gz" -) -FILES=( - "news.2007.de.shuffled.gz" - "news.2008.de.shuffled.gz" - "news.2009.de.shuffled.gz" - "news.2010.de.shuffled.gz" - "news.2011.de.shuffled.gz" - "news.2012.de.shuffled.gz" - "news.2013.de.shuffled.gz" - "news.2014.de.shuffled.v2.gz" - "news.2015.de.shuffled.gz" - "news.2016.de.shuffled.gz" - "news.2017.de.shuffled.deduped.gz" -) - - -cd $orig -for ((i=0;i<${#URLS[@]};++i)); do - file=${FILES[i]} - if [ -f $file ]; then - echo "$file already exists, skipping download" - else - url=${URLS[i]} - wget "$url" - fi -done -cd .. - - -if [ -f $tmp/monolingual.${SUBSAMPLE_SIZE}.${LANG} ]; then - echo "found monolingual sample, skipping shuffle/sample/tokenize" -else - gzip -c -d -k $(for FILE in "${FILES[@]}"; do echo $orig/$FILE; done) \ - | shuf -n $SUBSAMPLE_SIZE \ - | perl $NORM_PUNC $LANG \ - | perl $REM_NON_PRINT_CHAR \ - | perl $TOKENIZER -threads 8 -a -l $LANG \ - > $tmp/monolingual.${SUBSAMPLE_SIZE}.${LANG} -fi - - -if [ -f $tmp/bpe.monolingual.${SUBSAMPLE_SIZE}.${LANG} ]; then - echo "found BPE monolingual sample, skipping BPE step" -else - python $BPEROOT/apply_bpe.py -c $BPE_CODE \ - < $tmp/monolingual.${SUBSAMPLE_SIZE}.${LANG} \ - > $tmp/bpe.monolingual.${SUBSAMPLE_SIZE}.${LANG} -fi - - -if [ -f $tmp/bpe.monolingual.dedup.${SUBSAMPLE_SIZE}.${LANG} ]; then - echo "found deduplicated monolingual sample, skipping deduplication step" -else - python deduplicate_lines.py $tmp/bpe.monolingual.${SUBSAMPLE_SIZE}.${LANG} \ - > $tmp/bpe.monolingual.dedup.${SUBSAMPLE_SIZE}.${LANG} -fi - - -if [ -f $OUTDIR/bpe.monolingual.dedup.00.de ]; then - echo "found sharded data, skipping sharding step" -else - split --lines 1000000 --numeric-suffixes \ - --additional-suffix .${LANG} \ - $tmp/bpe.monolingual.dedup.${SUBSAMPLE_SIZE}.${LANG} \ - $OUTDIR/bpe.monolingual.dedup. -fi diff --git a/fairseq/examples/backtranslation/prepare-wmt18en2de.sh b/fairseq/examples/backtranslation/prepare-wmt18en2de.sh deleted file mode 100644 index f6fd275307db50ca84c299440ae02dce49064030..0000000000000000000000000000000000000000 --- a/fairseq/examples/backtranslation/prepare-wmt18en2de.sh +++ /dev/null @@ -1,135 +0,0 @@ -#!/bin/bash -# Adapted from https://github.com/facebookresearch/MIXER/blob/master/prepareData.sh - -echo 'Cloning Moses github repository (for tokenization scripts)...' -git clone https://github.com/moses-smt/mosesdecoder.git - -echo 'Cloning Subword NMT repository (for BPE pre-processing)...' -git clone https://github.com/rsennrich/subword-nmt.git - -SCRIPTS=mosesdecoder/scripts -TOKENIZER=$SCRIPTS/tokenizer/tokenizer.perl -CLEAN=$SCRIPTS/training/clean-corpus-n.perl -NORM_PUNC=$SCRIPTS/tokenizer/normalize-punctuation.perl -REM_NON_PRINT_CHAR=$SCRIPTS/tokenizer/remove-non-printing-char.perl -BPEROOT=subword-nmt/subword_nmt -BPE_TOKENS=32000 - -URLS=( - "http://statmt.org/wmt13/training-parallel-europarl-v7.tgz" - "http://statmt.org/wmt13/training-parallel-commoncrawl.tgz" - "http://data.statmt.org/wmt18/translation-task/training-parallel-nc-v13.tgz" - "http://data.statmt.org/wmt18/translation-task/rapid2016.tgz" - "http://data.statmt.org/wmt17/translation-task/dev.tgz" - "http://statmt.org/wmt14/test-full.tgz" -) -FILES=( - "training-parallel-europarl-v7.tgz" - "training-parallel-commoncrawl.tgz" - "training-parallel-nc-v13.tgz" - "rapid2016.tgz" - "dev.tgz" - "test-full.tgz" -) -CORPORA=( - "training/europarl-v7.de-en" - "commoncrawl.de-en" - "training-parallel-nc-v13/news-commentary-v13.de-en" - "rapid2016.de-en" -) - -if [ ! -d "$SCRIPTS" ]; then - echo "Please set SCRIPTS variable correctly to point to Moses scripts." - exit 1 -fi - -OUTDIR=wmt18_en_de - -src=en -tgt=de -lang=en-de -prep=$OUTDIR -tmp=$prep/tmp -orig=orig - -mkdir -p $orig $tmp $prep - -cd $orig - -for ((i=0;i<${#URLS[@]};++i)); do - file=${FILES[i]} - if [ -f $file ]; then - echo "$file already exists, skipping download" - else - url=${URLS[i]} - wget "$url" - if [ -f $file ]; then - echo "$url successfully downloaded." - else - echo "$url not successfully downloaded." - exit 1 - fi - if [ ${file: -4} == ".tgz" ]; then - tar zxvf $file - elif [ ${file: -4} == ".tar" ]; then - tar xvf $file - fi - fi -done -cd .. - -echo "pre-processing train data..." -for l in $src $tgt; do - rm $tmp/train.tags.$lang.tok.$l - for f in "${CORPORA[@]}"; do - cat $orig/$f.$l | \ - perl $NORM_PUNC $l | \ - perl $REM_NON_PRINT_CHAR | \ - perl $TOKENIZER -threads 8 -a -l $l >> $tmp/train.tags.$lang.tok.$l - done -done - -echo "pre-processing test data..." -for l in $src $tgt; do - if [ "$l" == "$src" ]; then - t="src" - else - t="ref" - fi - grep '
([Gehring et al., 2017](https://arxiv.org/abs/1705.03122)) | [WMT14 English-French](http://statmt.org/wmt14/translation-task.html#Download) | [download (.tar.bz2)](https://dl.fbaipublicfiles.com/fairseq/models/wmt14.v2.en-fr.fconv-py.tar.bz2) | newstest2014:
[download (.tar.bz2)](https://dl.fbaipublicfiles.com/fairseq/data/wmt14.v2.en-fr.newstest2014.tar.bz2)
newstest2012/2013:
[download (.tar.bz2)](https://dl.fbaipublicfiles.com/fairseq/data/wmt14.v2.en-fr.ntst1213.tar.bz2) -Convolutional
([Gehring et al., 2017](https://arxiv.org/abs/1705.03122)) | [WMT14 English-German](http://statmt.org/wmt14/translation-task.html#Download) | [download (.tar.bz2)](https://dl.fbaipublicfiles.com/fairseq/models/wmt14.en-de.fconv-py.tar.bz2) | newstest2014:
[download (.tar.bz2)](https://dl.fbaipublicfiles.com/fairseq/data/wmt14.en-de.newstest2014.tar.bz2) -Convolutional
([Gehring et al., 2017](https://arxiv.org/abs/1705.03122)) | [WMT17 English-German](http://statmt.org/wmt17/translation-task.html#Download) | [download (.tar.bz2)](https://dl.fbaipublicfiles.com/fairseq/models/wmt17.v2.en-de.fconv-py.tar.bz2) | newstest2014:
[download (.tar.bz2)](https://dl.fbaipublicfiles.com/fairseq/data/wmt17.v2.en-de.newstest2014.tar.bz2) - -## Example usage - -See the [translation README](../translation/README.md) for instructions on reproducing results for WMT'14 En-De and -WMT'14 En-Fr using the `fconv_wmt_en_de` and `fconv_wmt_en_fr` model architectures. - -## Citation - -```bibtex -@inproceedings{gehring2017convs2s, - title = {Convolutional Sequence to Sequence Learning}, - author = {Gehring, Jonas, and Auli, Michael and Grangier, David and Yarats, Denis and Dauphin, Yann N}, - booktitle = {Proc. of ICML}, - year = 2017, -} -``` diff --git a/fairseq/examples/criss/README.md b/fairseq/examples/criss/README.md deleted file mode 100644 index 4689ed7c10497a5100b28fe6d6801a7c089da569..0000000000000000000000000000000000000000 --- a/fairseq/examples/criss/README.md +++ /dev/null @@ -1,61 +0,0 @@ -# Cross-lingual Retrieval for Iterative Self-Supervised Training - -https://arxiv.org/pdf/2006.09526.pdf - -## Introduction - -CRISS is a multilingual sequence-to-sequnce pretraining method where mining and training processes are applied iteratively, improving cross-lingual alignment and translation ability at the same time. - -## Requirements: - -* faiss: https://github.com/facebookresearch/faiss -* mosesdecoder: https://github.com/moses-smt/mosesdecoder -* flores: https://github.com/facebookresearch/flores -* LASER: https://github.com/facebookresearch/LASER - -## Unsupervised Machine Translation -##### 1. Download and decompress CRISS checkpoints -``` -cd examples/criss -wget https://dl.fbaipublicfiles.com/criss/criss_3rd_checkpoints.tar.gz -tar -xf criss_checkpoints.tar.gz -``` -##### 2. Download and preprocess Flores test dataset -Make sure to run all scripts from examples/criss directory -``` -bash download_and_preprocess_flores_test.sh -``` - -##### 3. Run Evaluation on Sinhala-English -``` -bash unsupervised_mt/eval.sh -``` - -## Sentence Retrieval -##### 1. Download and preprocess Tatoeba dataset -``` -bash download_and_preprocess_tatoeba.sh -``` - -##### 2. Run Sentence Retrieval on Tatoeba Kazakh-English -``` -bash sentence_retrieval/sentence_retrieval_tatoeba.sh -``` - -## Mining -##### 1. Install faiss -Follow instructions on https://github.com/facebookresearch/faiss/blob/master/INSTALL.md -##### 2. Mine pseudo-parallel data between Kazakh and English -``` -bash mining/mine_example.sh -``` - -## Citation -```bibtex -@article{tran2020cross, - title={Cross-lingual retrieval for iterative self-supervised training}, - author={Tran, Chau and Tang, Yuqing and Li, Xian and Gu, Jiatao}, - journal={arXiv preprint arXiv:2006.09526}, - year={2020} -} -``` diff --git a/fairseq/examples/criss/download_and_preprocess_flores_test.sh b/fairseq/examples/criss/download_and_preprocess_flores_test.sh deleted file mode 100644 index ed4b390fbdee3991efeb298050e12065d7fe605b..0000000000000000000000000000000000000000 --- a/fairseq/examples/criss/download_and_preprocess_flores_test.sh +++ /dev/null @@ -1,64 +0,0 @@ -#!/bin/bash -# Copyright (c) Facebook, Inc. and its affiliates. -# All rights reserved. -# -# This source code is licensed under the license found in the -# LICENSE file in the root directory of this source tree. - -SPM_ENCODE=flores/scripts/spm_encode.py -DATA=data_tmp -SPM_MODEL=criss_checkpoints/sentence.bpe.model -DICT=criss_checkpoints/dict.txt - -download_data() { - CORPORA=$1 - URL=$2 - - if [ -f $CORPORA ]; then - echo "$CORPORA already exists, skipping download" - else - echo "Downloading $URL" - wget $URL -O $CORPORA --no-check-certificate || rm -f $CORPORA - if [ -f $CORPORA ]; then - echo "$URL successfully downloaded." - else - echo "$URL not successfully downloaded." - rm -f $CORPORA - fi - fi -} - -if [[ -f flores ]]; then - echo "flores already cloned" -else - git clone https://github.com/facebookresearch/flores -fi - -mkdir -p $DATA -download_data $DATA/wikipedia_en_ne_si_test_sets.tgz "https://github.com/facebookresearch/flores/raw/master/data/wikipedia_en_ne_si_test_sets.tgz" -pushd $DATA -pwd -tar -vxf wikipedia_en_ne_si_test_sets.tgz -popd - - -for lang in ne_NP si_LK; do - datadir=$DATA/${lang}-en_XX-flores - rm -rf $datadir - mkdir -p $datadir - TEST_PREFIX=$DATA/wikipedia_en_ne_si_test_sets/wikipedia.test - python $SPM_ENCODE \ - --model ${SPM_MODEL} \ - --output_format=piece \ - --inputs ${TEST_PREFIX}.${lang:0:2}-en.${lang:0:2} ${TEST_PREFIX}.${lang:0:2}-en.en \ - --outputs $datadir/test.bpe.${lang}-en_XX.${lang} $datadir/test.bpe.${lang}-en_XX.en_XX - - # binarize data - fairseq-preprocess \ - --source-lang ${lang} --target-lang en_XX \ - --testpref $datadir/test.bpe.${lang}-en_XX \ - --destdir $datadir \ - --srcdict ${DICT} \ - --joined-dictionary \ - --workers 4 -done diff --git a/fairseq/examples/criss/download_and_preprocess_tatoeba.sh b/fairseq/examples/criss/download_and_preprocess_tatoeba.sh deleted file mode 100644 index 7ed64f017d5e62695ba73745c840507b994abc0f..0000000000000000000000000000000000000000 --- a/fairseq/examples/criss/download_and_preprocess_tatoeba.sh +++ /dev/null @@ -1,46 +0,0 @@ -#!/bin/bash -# Copyright (c) Facebook, Inc. and its affiliates. -# All rights reserved. -# -# This source code is licensed under the license found in the -# LICENSE file in the root directory of this source tree. - -SPM_ENCODE=flores/scripts/spm_encode.py -DATA=data_tmp -SPM_MODEL=criss_checkpoints/sentence.bpe.model -DICT=criss_checkpoints/dict.txt - -if [[ -f flores ]]; then - echo "flores already cloned" -else - git clone https://github.com/facebookresearch/flores -fi -if [[ -f LASER ]]; then - echo "LASER already cloned" -else - git clone https://github.com/facebookresearch/LASER -fi -mkdir -p data_tmp -declare -A lang_tatoeba_map=( ["ar_AR"]="ara" ["de_DE"]="deu" ["es_XX"]="spa" ["et_EE"]="est" ["fi_FI"]="fin" ["fr_XX"]="fra" ["hi_IN"]="hin" ["it_IT"]="ita" ["ja_XX"]="jpn" ["ko_KR"]="kor" ["kk_KZ"]="kaz" ["nl_XX"]="nld" ["ru_RU"]="rus" ["tr_TR"]="tur" ["vi_VN"]="vie" ["zh_CN"]="cmn") -for lang in ar_AR de_DE es_XX et_EE fi_FI fr_XX hi_IN it_IT ja_XX kk_KZ ko_KR nl_XX ru_RU tr_TR vi_VN zh_CN; do - lang_tatoeba=${lang_tatoeba_map[$lang]} - echo $lang_tatoeba - datadir=$DATA/${lang}-en_XX-tatoeba - rm -rf $datadir - mkdir -p $datadir - TEST_PREFIX=LASER/data/tatoeba/v1/tatoeba - python $SPM_ENCODE \ - --model ${SPM_MODEL} \ - --output_format=piece \ - --inputs ${TEST_PREFIX}.${lang_tatoeba}-eng.${lang_tatoeba} ${TEST_PREFIX}.${lang_tatoeba}-eng.eng \ - --outputs $datadir/test.bpe.${lang}-en_XX.${lang} $datadir/test.bpe.${lang}-en_XX.en_XX - - # binarize data - fairseq-preprocess \ - --source-lang ${lang} --target-lang en_XX \ - --testpref $datadir/test.bpe.${lang}-en_XX \ - --destdir $datadir \ - --srcdict ${DICT} \ - --joined-dictionary \ - --workers 4 -done diff --git a/fairseq/examples/criss/mining/mine.py b/fairseq/examples/criss/mining/mine.py deleted file mode 100644 index c872da196fe0df776622365748ad7963fee1f0a0..0000000000000000000000000000000000000000 --- a/fairseq/examples/criss/mining/mine.py +++ /dev/null @@ -1,240 +0,0 @@ -#!/usr/bin/env python3 -u -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. -import argparse -import glob -from subprocess import check_call - -try: - import faiss - - has_faiss = True -except ImportError: - has_faiss = False -import numpy as np - - -GB = 1024 * 1024 * 1024 - - -def call(cmd): - print(cmd) - check_call(cmd, shell=True) - - -def get_batches(directory, lang, prefix="all_avg_pool"): - print(f"Finding in {directory}/{prefix}.{lang}*") - files = glob.glob(f"{directory}/{prefix}.{lang}*") - emb_files = [] - txt_files = [] - for emb_fi in files: - emb_files.append(emb_fi) - txt_fi = emb_fi.replace(prefix, "sentences") - txt_files.append(txt_fi) - return emb_files, txt_files - - -def load_batch(emb_file, dim): - embeddings = np.fromfile(emb_file, dtype=np.float32) - num_rows = int(embeddings.shape[0] / dim) - embeddings = embeddings.reshape((num_rows, dim)) - faiss.normalize_L2(embeddings) - return embeddings - - -def knnGPU_sharded(x_batches_f, y_batches_f, dim, k, direction="x2y"): - if not has_faiss: - raise ImportError("Please install Faiss") - sims = [] - inds = [] - xfrom = 0 - xto = 0 - for x_batch_f in x_batches_f: - yfrom = 0 - yto = 0 - x_batch = load_batch(x_batch_f, dim) - xto = xfrom + x_batch.shape[0] - bsims, binds = [], [] - for y_batch_f in y_batches_f: - y_batch = load_batch(y_batch_f, dim) - neighbor_size = min(k, y_batch.shape[0]) - yto = yfrom + y_batch.shape[0] - print("{}-{} -> {}-{}".format(xfrom, xto, yfrom, yto)) - idx = faiss.IndexFlatIP(dim) - idx = faiss.index_cpu_to_all_gpus(idx) - idx.add(y_batch) - bsim, bind = idx.search(x_batch, neighbor_size) - - bsims.append(bsim) - binds.append(bind + yfrom) - yfrom += y_batch.shape[0] - del idx - del y_batch - bsims = np.concatenate(bsims, axis=1) - binds = np.concatenate(binds, axis=1) - aux = np.argsort(-bsims, axis=1) - sim_batch = np.zeros((x_batch.shape[0], k), dtype=np.float32) - ind_batch = np.zeros((x_batch.shape[0], k), dtype=np.int64) - for i in range(x_batch.shape[0]): - for j in range(k): - sim_batch[i, j] = bsims[i, aux[i, j]] - ind_batch[i, j] = binds[i, aux[i, j]] - sims.append(sim_batch) - inds.append(ind_batch) - xfrom += x_batch.shape[0] - del x_batch - sim = np.concatenate(sims, axis=0) - ind = np.concatenate(inds, axis=0) - return sim, ind - - -def score(sim, fwd_mean, bwd_mean, margin): - return margin(sim, (fwd_mean + bwd_mean) / 2) - - -def score_candidates( - sim_mat, candidate_inds, fwd_mean, bwd_mean, margin, verbose=False -): - print(" - scoring {:d} candidates".format(sim_mat.shape[0])) - scores = np.zeros(candidate_inds.shape) - for i in range(scores.shape[0]): - for j in range(scores.shape[1]): - k = int(candidate_inds[i, j]) - scores[i, j] = score(sim_mat[i, j], fwd_mean[i], bwd_mean[k], margin) - return scores - - -def load_text(files): - all_sentences = [] - for fi in files: - with open(fi) as sentence_fi: - for line in sentence_fi: - all_sentences.append(line.strip()) - print(f"Read {len(all_sentences)} sentences") - return all_sentences - - -if __name__ == "__main__": - parser = argparse.ArgumentParser(description="Mine bitext") - parser.add_argument("--src-lang", help="Source language") - parser.add_argument("--tgt-lang", help="Target language") - parser.add_argument( - "--dict-path", help="Path to dictionary file", default="dict.txt" - ) - parser.add_argument( - "--spm-path", help="Path to SPM model file", default="sentence.bpe.model" - ) - parser.add_argument("--dim", type=int, default=1024, help="Embedding dimension") - parser.add_argument("--mem", type=int, default=5, help="Memory in GB") - parser.add_argument("--src-dir", help="Source directory") - parser.add_argument("--tgt-dir", help="Target directory") - parser.add_argument("--output", help="Output path") - parser.add_argument( - "--neighborhood", type=int, default=4, help="Embedding dimension" - ) - parser.add_argument( - "--threshold", type=float, default=1.06, help="Threshold on mined bitext" - ) - parser.add_argument( - "--valid-size", - type=int, - default=2000, - help="Number of sentences used for validation set", - ) - parser.add_argument( - "--min-count", - type=int, - default=50000, - help="Min num sentences used for each language", - ) - args = parser.parse_args() - - x_batches_f, x_sents_f = get_batches(args.src_dir, args.src_lang) - y_batches_f, y_sents_f = get_batches(args.tgt_dir, args.tgt_lang) - margin = lambda a, b: a / b - y2x_sim, y2x_ind = knnGPU_sharded( - y_batches_f, x_batches_f, args.dim, args.neighborhood, direction="y2x" - ) - x2y_sim, x2y_ind = knnGPU_sharded( - x_batches_f, y_batches_f, args.dim, args.neighborhood, direction="x2y" - ) - - x2y_mean = x2y_sim.mean(axis=1) - y2x_mean = y2x_sim.mean(axis=1) - fwd_scores = score_candidates(x2y_sim, x2y_ind, x2y_mean, y2x_mean, margin) - bwd_scores = score_candidates(y2x_sim, y2x_ind, y2x_mean, x2y_mean, margin) - fwd_best = x2y_ind[np.arange(x2y_sim.shape[0]), fwd_scores.argmax(axis=1)] - bwd_best = y2x_ind[np.arange(y2x_sim.shape[0]), bwd_scores.argmax(axis=1)] - indices = np.stack( - ( - np.concatenate((np.arange(x2y_ind.shape[0]), bwd_best)), - np.concatenate((fwd_best, np.arange(y2x_ind.shape[0]))), - ), - axis=1, - ) - scores = np.concatenate((fwd_scores.max(axis=1), bwd_scores.max(axis=1))) - - x_sentences = load_text(x_sents_f) - y_sentences = load_text(y_sents_f) - - threshold = args.threshold - min_count = args.min_count - seen_src, seen_trg = set(), set() - directory = args.output - call(f"mkdir -p {directory}") - src_out = open( - f"{directory}/all.{args.src_lang}", - mode="w", - encoding="utf-8", - errors="surrogateescape", - ) - tgt_out = open( - f"{directory}/all.{args.tgt_lang}", - mode="w", - encoding="utf-8", - errors="surrogateescape", - ) - scores_out = open( - f"{directory}/all.scores", mode="w", encoding="utf-8", errors="surrogateescape" - ) - count = 0 - for i in np.argsort(-scores): - src_ind, trg_ind = indices[i] - if src_ind not in seen_src and trg_ind not in seen_trg: - seen_src.add(src_ind) - seen_trg.add(trg_ind) - if scores[i] > threshold or count < min_count: - if x_sentences[src_ind]: - print(scores[i], file=scores_out) - print(x_sentences[src_ind], file=src_out) - print(y_sentences[trg_ind], file=tgt_out) - count += 1 - else: - print(f"Ignoring sentence: {x_sentences[src_ind]}") - src_out.close() - tgt_out.close() - scores_out.close() - - print(f"Found {count} pairs for threshold={threshold}") - with open(f"{directory}/all.{args.src_lang}") as all_s, open( - f"{directory}/all.{args.tgt_lang}" - ) as all_t, open(f"{directory}/valid.{args.src_lang}", "w") as valid_s, open( - f"{directory}/valid.{args.tgt_lang}", "w" - ) as valid_t, open( - f"{directory}/train.{args.src_lang}", "w" - ) as train_s, open( - f"{directory}/train.{args.tgt_lang}", "w" - ) as train_t: - count = 0 - for s_line, t_line in zip(all_s, all_t): - s_line = s_line.split("\t")[1] - t_line = t_line.split("\t")[1] - if count >= args.valid_size: - train_s.write(s_line) - train_t.write(t_line) - else: - valid_s.write(s_line) - valid_t.write(t_line) - count += 1 diff --git a/fairseq/examples/criss/mining/mine_example.sh b/fairseq/examples/criss/mining/mine_example.sh deleted file mode 100644 index ace995ac44665f99d904b6a89d7fbbce24103afe..0000000000000000000000000000000000000000 --- a/fairseq/examples/criss/mining/mine_example.sh +++ /dev/null @@ -1,103 +0,0 @@ -#!/bin/bash -# Copyright (c) Facebook, Inc. and its affiliates. -# All rights reserved. -# -# This source code is licensed under the license found in the -# LICENSE file in the root directory of this source tree. -# -source_lang=kk_KZ -target_lang=en_XX -MODEL=criss_checkpoints/criss.3rd.pt -SPM=criss_checkpoints/sentence.bpe.model -SPLIT=test -LANG_DICT=criss_checkpoints/lang_dict.txt -SPM_ENCODE=flores/scripts/spm_encode.py -SAVE_ENCODER=save_encoder.py -ENCODER_SAVE_ROOT=sentence_embeddings/$MODEL -DICT=criss_checkpoints/dict.txt -THRESHOLD=1.02 -MIN_COUNT=500 - -DATA_DIR=data_tmp -SAVE_DIR=mining/${source_lang}_${target_lang}_mined -ENCODER_SAVE_DIR=${ENCODER_SAVE_ROOT}/${source_lang}-${target_lang} -INPUT_DIR=$DATA_DIR/${source_lang}-${target_lang}-tatoeba - -mkdir -p $ENCODER_SAVE_DIR/${target_lang} -mkdir -p $ENCODER_SAVE_DIR/${source_lang} -mkdir -p $SAVE_DIR - -## Save encoder outputs - -# Save encoder outputs for source sentences -python $SAVE_ENCODER \ - ${INPUT_DIR} \ - --path ${MODEL} \ - --task translation_multi_simple_epoch \ - --lang-pairs ${source_lang}-${target_lang} \ - --lang-dict ${LANG_DICT} \ - --gen-subset ${SPLIT} \ - --bpe 'sentencepiece' \ - -s ${source_lang} -t ${target_lang} \ - --sentencepiece-model ${SPM} \ - --remove-bpe 'sentencepiece' \ - --beam 1 \ - --lang-tok-style mbart \ - --encoder-save-dir ${ENCODER_SAVE_DIR}/${source_lang} - -## Save encoder outputs for target sentences -python $SAVE_ENCODER \ - ${INPUT_DIR} \ - --path ${MODEL} \ - --lang-pairs ${source_lang}-${target_lang} \ - --lang-dict ${LANG_DICT} \ - --task translation_multi_simple_epoch \ - --gen-subset ${SPLIT} \ - --bpe 'sentencepiece' \ - -t ${source_lang} -s ${target_lang} \ - --sentencepiece-model ${SPM} \ - --remove-bpe 'sentencepiece' \ - --beam 1 \ - --lang-tok-style mbart \ - --encoder-save-dir ${ENCODER_SAVE_DIR}/${target_lang} - -## Mining -python mining/mine.py \ - --src-lang ${source_lang} \ - --tgt-lang ${target_lang} \ - --dim 1024 \ - --mem 10 \ - --neighborhood 4 \ - --src-dir ${ENCODER_SAVE_DIR}/${source_lang} \ - --tgt-dir ${ENCODER_SAVE_DIR}/${target_lang} \ - --output $SAVE_DIR \ - --threshold ${THRESHOLD} \ - --min-count ${MIN_COUNT} \ - --valid-size 100 \ - --dict-path ${DICT} \ - --spm-path ${SPM} \ - - -## Process and binarize mined data -python $SPM_ENCODE \ - --model ${SPM} \ - --output_format=piece \ - --inputs mining/${source_lang}_${target_lang}_mined/train.${source_lang} mining/${source_lang}_${target_lang}_mined/train.${target_lang} \ - --outputs mining/${source_lang}_${target_lang}_mined/train.bpe.${source_lang} mining/${source_lang}_${target_lang}_mined/train.bpe.${target_lang} - -python $SPM_ENCODE \ - --model ${SPM} \ - --output_format=piece \ - --inputs mining/${source_lang}_${target_lang}_mined/valid.${source_lang} mining/${source_lang}_${target_lang}_mined/valid.${target_lang} \ - --outputs mining/${source_lang}_${target_lang}_mined/valid.bpe.${source_lang} mining/${source_lang}_${target_lang}_mined/valid.bpe.${target_lang} - - -fairseq-preprocess \ - --source-lang ${source_lang} \ - --target-lang ${target_lang} \ - --trainpref mining/${source_lang}_${target_lang}_mined/train.bpe \ - --validpref mining/${source_lang}_${target_lang}_mined/valid.bpe \ - --destdir mining/${source_lang}_${target_lang}_mined \ - --srcdict ${DICT} \ - --joined-dictionary \ - --workers 8 diff --git a/fairseq/examples/criss/save_encoder.py b/fairseq/examples/criss/save_encoder.py deleted file mode 100644 index 24a842e4092663c79c92a299fa85747b7c0bed64..0000000000000000000000000000000000000000 --- a/fairseq/examples/criss/save_encoder.py +++ /dev/null @@ -1,214 +0,0 @@ -#!/usr/bin/env python3 -u -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. -""" -Translate pre-processed data with a trained model. -""" - -import numpy as np -import torch -from fairseq import checkpoint_utils, options, progress_bar, tasks, utils -from fairseq.sequence_generator import EnsembleModel -from fairseq.utils import safe_hasattr - - -def get_avg_pool( - models, sample, prefix_tokens, src_dict, remove_bpe, has_langtok=False -): - model = EnsembleModel(models) - - # model.forward normally channels prev_output_tokens into the decoder - # separately, but SequenceGenerator directly calls model.encoder - encoder_input = { - k: v for k, v in sample["net_input"].items() if k != "prev_output_tokens" - } - - # compute the encoder output for each beam - encoder_outs = model.forward_encoder(encoder_input) - np_encoder_outs = encoder_outs[0].encoder_out.cpu().numpy().astype(np.float32) - encoder_mask = 1 - encoder_outs[0].encoder_padding_mask.cpu().numpy().astype( - np.float32 - ) - encoder_mask = np.expand_dims(encoder_mask.T, axis=2) - if has_langtok: - encoder_mask = encoder_mask[1:, :, :] - np_encoder_outs = np_encoder_outs[1, :, :] - masked_encoder_outs = encoder_mask * np_encoder_outs - avg_pool = (masked_encoder_outs / encoder_mask.sum(axis=0)).sum(axis=0) - return avg_pool - - -def main(args): - assert args.path is not None, "--path required for generation!" - assert ( - not args.sampling or args.nbest == args.beam - ), "--sampling requires --nbest to be equal to --beam" - assert ( - args.replace_unk is None or args.raw_text - ), "--replace-unk requires a raw text dataset (--raw-text)" - - args.beam = 1 - utils.import_user_module(args) - - if args.max_tokens is None: - args.max_tokens = 12000 - print(args) - use_cuda = torch.cuda.is_available() and not args.cpu - - # Load dataset splits - task = tasks.setup_task(args) - task.load_dataset(args.gen_subset) - - # Set dictionaries - try: - src_dict = getattr(task, "source_dictionary", None) - except NotImplementedError: - src_dict = None - tgt_dict = task.target_dictionary - - # Load ensemble - print("| loading model(s) from {}".format(args.path)) - models, _model_args = checkpoint_utils.load_model_ensemble( - args.path.split(":"), - arg_overrides=eval(args.model_overrides), - task=task, - ) - - # Optimize ensemble for generation - for model in models: - model.make_generation_fast_( - beamable_mm_beam_size=None if args.no_beamable_mm else args.beam, - need_attn=args.print_alignment, - ) - if args.fp16: - model.half() - if use_cuda: - model.cuda() - - # Load alignment dictionary for unknown word replacement - # (None if no unknown word replacement, empty if no path to align dictionary) - align_dict = utils.load_align_dict(args.replace_unk) - - # Load dataset (possibly sharded) - itr = task.get_batch_iterator( - dataset=task.dataset(args.gen_subset), - max_tokens=args.max_tokens, - max_positions=utils.resolve_max_positions( - task.max_positions(), - ), - ignore_invalid_inputs=args.skip_invalid_size_inputs_valid_test, - required_batch_size_multiple=args.required_batch_size_multiple, - num_shards=args.num_shards, - shard_id=args.shard_id, - num_workers=args.num_workers, - ).next_epoch_itr(shuffle=False) - - num_sentences = 0 - source_sentences = [] - shard_id = 0 - all_avg_pool = None - encoder_has_langtok = ( - safe_hasattr(task.args, "encoder_langtok") - and task.args.encoder_langtok is not None - and safe_hasattr(task.args, "lang_tok_replacing_bos_eos") - and not task.args.lang_tok_replacing_bos_eos - ) - with progress_bar.build_progress_bar(args, itr) as t: - for sample in t: - if sample is None: - print("Skipping None") - continue - sample = utils.move_to_cuda(sample) if use_cuda else sample - if "net_input" not in sample: - continue - - prefix_tokens = None - if args.prefix_size > 0: - prefix_tokens = sample["target"][:, : args.prefix_size] - - with torch.no_grad(): - avg_pool = get_avg_pool( - models, - sample, - prefix_tokens, - src_dict, - args.post_process, - has_langtok=encoder_has_langtok, - ) - if all_avg_pool is not None: - all_avg_pool = np.concatenate((all_avg_pool, avg_pool)) - else: - all_avg_pool = avg_pool - - if not isinstance(sample["id"], list): - sample_ids = sample["id"].tolist() - else: - sample_ids = sample["id"] - for i, sample_id in enumerate(sample_ids): - # Remove padding - src_tokens = utils.strip_pad( - sample["net_input"]["src_tokens"][i, :], tgt_dict.pad() - ) - - # Either retrieve the original sentences or regenerate them from tokens. - if align_dict is not None: - src_str = task.dataset(args.gen_subset).src.get_original_text( - sample_id - ) - else: - if src_dict is not None: - src_str = src_dict.string(src_tokens, args.post_process) - else: - src_str = "" - - if not args.quiet: - if src_dict is not None: - print("S-{}\t{}".format(sample_id, src_str)) - - source_sentences.append(f"{sample_id}\t{src_str}") - - num_sentences += sample["nsentences"] - if all_avg_pool.shape[0] >= 1000000: - with open( - f"{args.encoder_save_dir}/all_avg_pool.{args.source_lang}.{shard_id}", - "w", - ) as avg_pool_file: - all_avg_pool.tofile(avg_pool_file) - with open( - f"{args.encoder_save_dir}/sentences.{args.source_lang}.{shard_id}", - "w", - ) as sentence_file: - sentence_file.writelines(f"{line}\n" for line in source_sentences) - all_avg_pool = None - source_sentences = [] - shard_id += 1 - - if all_avg_pool is not None: - with open( - f"{args.encoder_save_dir}/all_avg_pool.{args.source_lang}.{shard_id}", "w" - ) as avg_pool_file: - all_avg_pool.tofile(avg_pool_file) - with open( - f"{args.encoder_save_dir}/sentences.{args.source_lang}.{shard_id}", "w" - ) as sentence_file: - sentence_file.writelines(f"{line}\n" for line in source_sentences) - return None - - -def cli_main(): - parser = options.get_generation_parser() - parser.add_argument( - "--encoder-save-dir", - default="", - type=str, - metavar="N", - help="directory to save encoder outputs", - ) - args = options.parse_args_and_arch(parser) - main(args) - - -if __name__ == "__main__": - cli_main() diff --git a/fairseq/examples/criss/sentence_retrieval/encoder_analysis.py b/fairseq/examples/criss/sentence_retrieval/encoder_analysis.py deleted file mode 100644 index b41bfbe38789ba14e6a5ea938c75d761424c00ab..0000000000000000000000000000000000000000 --- a/fairseq/examples/criss/sentence_retrieval/encoder_analysis.py +++ /dev/null @@ -1,92 +0,0 @@ -#!/usr/bin/env python3 -u -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. -import argparse -import glob - -import numpy as np - - -DIM = 1024 - - -def compute_dist(source_embs, target_embs, k=5, return_sim_mat=False): - target_ids = [tid for tid in target_embs] - source_mat = np.stack(source_embs.values(), axis=0) - normalized_source_mat = source_mat / np.linalg.norm( - source_mat, axis=1, keepdims=True - ) - target_mat = np.stack(target_embs.values(), axis=0) - normalized_target_mat = target_mat / np.linalg.norm( - target_mat, axis=1, keepdims=True - ) - sim_mat = normalized_source_mat.dot(normalized_target_mat.T) - if return_sim_mat: - return sim_mat - neighbors_map = {} - for i, sentence_id in enumerate(source_embs): - idx = np.argsort(sim_mat[i, :])[::-1][:k] - neighbors_map[sentence_id] = [target_ids[tid] for tid in idx] - return neighbors_map - - -def load_embeddings(directory, LANGS): - sentence_embeddings = {} - sentence_texts = {} - for lang in LANGS: - sentence_embeddings[lang] = {} - sentence_texts[lang] = {} - lang_dir = f"{directory}/{lang}" - embedding_files = glob.glob(f"{lang_dir}/all_avg_pool.{lang}.*") - for embed_file in embedding_files: - shard_id = embed_file.split(".")[-1] - embeddings = np.fromfile(embed_file, dtype=np.float32) - num_rows = embeddings.shape[0] // DIM - embeddings = embeddings.reshape((num_rows, DIM)) - - with open(f"{lang_dir}/sentences.{lang}.{shard_id}") as sentence_file: - for idx, line in enumerate(sentence_file): - sentence_id, sentence = line.strip().split("\t") - sentence_texts[lang][sentence_id] = sentence - sentence_embeddings[lang][sentence_id] = embeddings[idx, :] - - return sentence_embeddings, sentence_texts - - -def compute_accuracy(directory, LANGS): - sentence_embeddings, sentence_texts = load_embeddings(directory, LANGS) - - top_1_accuracy = {} - - top1_str = " ".join(LANGS) + "\n" - for source_lang in LANGS: - top_1_accuracy[source_lang] = {} - top1_str += f"{source_lang} " - for target_lang in LANGS: - top1 = 0 - top5 = 0 - neighbors_map = compute_dist( - sentence_embeddings[source_lang], sentence_embeddings[target_lang] - ) - for sentence_id, neighbors in neighbors_map.items(): - if sentence_id == neighbors[0]: - top1 += 1 - if sentence_id in neighbors[:5]: - top5 += 1 - n = len(sentence_embeddings[target_lang]) - top1_str += f"{top1/n} " - top1_str += "\n" - - print(top1_str) - print(top1_str, file=open(f"{directory}/accuracy", "w")) - - -if __name__ == "__main__": - parser = argparse.ArgumentParser(description="Analyze encoder outputs") - parser.add_argument("directory", help="Source language corpus") - parser.add_argument("--langs", help="List of langs") - args = parser.parse_args() - langs = args.langs.split(",") - compute_accuracy(args.directory, langs) diff --git a/fairseq/examples/criss/sentence_retrieval/sentence_retrieval_tatoeba.sh b/fairseq/examples/criss/sentence_retrieval/sentence_retrieval_tatoeba.sh deleted file mode 100644 index 0428d8bef9d426ac3e664cd281ce0b688f5f580f..0000000000000000000000000000000000000000 --- a/fairseq/examples/criss/sentence_retrieval/sentence_retrieval_tatoeba.sh +++ /dev/null @@ -1,59 +0,0 @@ -#!/bin/bash -# Copyright (c) Facebook, Inc. and its affiliates. -# All rights reserved. -# -# This source code is licensed under the license found in the -# LICENSE file in the root directory of this source tree. -# -source_lang=kk_KZ -target_lang=en_XX -MODEL=criss_checkpoints/criss.3rd.pt -SPM=criss_checkpoints/sentence.bpe.model -SPLIT=test -LANG_DICT=criss_checkpoints/lang_dict.txt -ENCODER_ANALYSIS=sentence_retrieval/encoder_analysis.py -SAVE_ENCODER=save_encoder.py -ENCODER_SAVE_ROOT=sentence_embeddings/$MODEL - - - -DATA_DIR=data_tmp -INPUT_DIR=$DATA_DIR/${source_lang}-${target_lang}-tatoeba -ENCODER_SAVE_DIR=${ENCODER_SAVE_ROOT}/${source_lang}-${target_lang} -mkdir -p $ENCODER_SAVE_DIR/${target_lang} -mkdir -p $ENCODER_SAVE_DIR/${source_lang} - -# Save encoder outputs for source sentences -python $SAVE_ENCODER \ - ${INPUT_DIR} \ - --path ${MODEL} \ - --task translation_multi_simple_epoch \ - --lang-dict ${LANG_DICT} \ - --gen-subset ${SPLIT} \ - --bpe 'sentencepiece' \ - --lang-pairs ${source_lang}-${target_lang} \ - -s ${source_lang} -t ${target_lang} \ - --sentencepiece-model ${SPM} \ - --remove-bpe 'sentencepiece' \ - --beam 1 \ - --lang-tok-style mbart \ - --encoder-save-dir ${ENCODER_SAVE_DIR}/${source_lang} - -# Save encoder outputs for target sentences -python $SAVE_ENCODER \ - ${INPUT_DIR} \ - --path ${MODEL} \ - --lang-dict ${LANG_DICT} \ - --task translation_multi_simple_epoch \ - --gen-subset ${SPLIT} \ - --bpe 'sentencepiece' \ - --lang-pairs ${target_lang}-${source_lang} \ - -t ${source_lang} -s ${target_lang} \ - --sentencepiece-model ${SPM} \ - --remove-bpe 'sentencepiece' \ - --beam 1 \ - --lang-tok-style mbart \ - --encoder-save-dir ${ENCODER_SAVE_DIR}/${target_lang} - -# Analyze sentence retrieval accuracy -python $ENCODER_ANALYSIS --langs "${source_lang},${target_lang}" ${ENCODER_SAVE_DIR} diff --git a/fairseq/examples/criss/unsupervised_mt/eval.sh b/fairseq/examples/criss/unsupervised_mt/eval.sh deleted file mode 100644 index 03b773ed5a522eb82186fea8ffbb6c557e14b6d3..0000000000000000000000000000000000000000 --- a/fairseq/examples/criss/unsupervised_mt/eval.sh +++ /dev/null @@ -1,37 +0,0 @@ -#!/bin/bash -# Copyright (c) Facebook, Inc. and its affiliates. -# All rights reserved. -# -# This source code is licensed under the license found in the -# LICENSE file in the root directory of this source tree. -# -SRC=si_LK -TGT=en_XX -MODEL=criss_checkpoints/criss.3rd.pt - -MULTIBLEU=mosesdecoder/scripts/generic/multi-bleu.perl -MOSES=mosesdecoder -REPLACE_UNICODE_PUNCT=$MOSES/scripts/tokenizer/replace-unicode-punctuation.perl -NORM_PUNC=$MOSES/scripts/tokenizer/normalize-punctuation.perl -REM_NON_PRINT_CHAR=$MOSES/scripts/tokenizer/remove-non-printing-char.perl -TOKENIZER=$MOSES/scripts/tokenizer/tokenizer.perl -GEN_TMP_DIR=gen_tmp -LANG_DICT=criss_checkpoints/lang_dict.txt - -if [ ! -d "mosesdecoder" ]; then - git clone https://github.com/moses-smt/mosesdecoder -fi -mkdir -p $GEN_TMP_DIR -fairseq-generate data_tmp/${SRC}-${TGT}-flores \ - --task translation_multi_simple_epoch \ - --max-tokens 2000 \ - --path ${MODEL} \ - --skip-invalid-size-inputs-valid-test \ - --beam 5 --lenpen 1.0 --gen-subset test \ - --remove-bpe=sentencepiece \ - --source-lang ${SRC} --target-lang ${TGT} \ - --decoder-langtok --lang-pairs 'en_XX-ar_AR,en_XX-de_DE,en_XX-es_XX,en_XX-fr_XX,en_XX-hi_IN,en_XX-it_IT,en_XX-ja_XX,en_XX-ko_KR,en_XX-nl_XX,en_XX-ru_RU,en_XX-zh_CN,en_XX-tr_TR,en_XX-vi_VN,en_XX-ro_RO,en_XX-my_MM,en_XX-ne_NP,en_XX-si_LK,en_XX-cs_CZ,en_XX-lt_LT,en_XX-kk_KZ,en_XX-gu_IN,en_XX-fi_FI,en_XX-et_EE,en_XX-lv_LV,ar_AR-en_XX,cs_CZ-en_XX,de_DE-en_XX,es_XX-en_XX,et_EE-en_XX,fi_FI-en_XX,fr_XX-en_XX,gu_IN-en_XX,hi_IN-en_XX,it_IT-en_XX,ja_XX-en_XX,kk_KZ-en_XX,ko_KR-en_XX,lt_LT-en_XX,lv_LV-en_XX,my_MM-en_XX,ne_NP-en_XX,nl_XX-en_XX,ro_RO-en_XX,ru_RU-en_XX,si_LK-en_XX,tr_TR-en_XX,vi_VN-en_XX,zh_CN-en_XX,ar_AR-es_XX,es_XX-ar_AR,ar_AR-hi_IN,hi_IN-ar_AR,ar_AR-zh_CN,zh_CN-ar_AR,cs_CZ-es_XX,es_XX-cs_CZ,cs_CZ-hi_IN,hi_IN-cs_CZ,cs_CZ-zh_CN,zh_CN-cs_CZ,de_DE-es_XX,es_XX-de_DE,de_DE-hi_IN,hi_IN-de_DE,de_DE-zh_CN,zh_CN-de_DE,es_XX-hi_IN,hi_IN-es_XX,es_XX-zh_CN,zh_CN-es_XX,et_EE-es_XX,es_XX-et_EE,et_EE-hi_IN,hi_IN-et_EE,et_EE-zh_CN,zh_CN-et_EE,fi_FI-es_XX,es_XX-fi_FI,fi_FI-hi_IN,hi_IN-fi_FI,fi_FI-zh_CN,zh_CN-fi_FI,fr_XX-es_XX,es_XX-fr_XX,fr_XX-hi_IN,hi_IN-fr_XX,fr_XX-zh_CN,zh_CN-fr_XX,gu_IN-es_XX,es_XX-gu_IN,gu_IN-hi_IN,hi_IN-gu_IN,gu_IN-zh_CN,zh_CN-gu_IN,hi_IN-zh_CN,zh_CN-hi_IN,it_IT-es_XX,es_XX-it_IT,it_IT-hi_IN,hi_IN-it_IT,it_IT-zh_CN,zh_CN-it_IT,ja_XX-es_XX,es_XX-ja_XX,ja_XX-hi_IN,hi_IN-ja_XX,ja_XX-zh_CN,zh_CN-ja_XX,kk_KZ-es_XX,es_XX-kk_KZ,kk_KZ-hi_IN,hi_IN-kk_KZ,kk_KZ-zh_CN,zh_CN-kk_KZ,ko_KR-es_XX,es_XX-ko_KR,ko_KR-hi_IN,hi_IN-ko_KR,ko_KR-zh_CN,zh_CN-ko_KR,lt_LT-es_XX,es_XX-lt_LT,lt_LT-hi_IN,hi_IN-lt_LT,lt_LT-zh_CN,zh_CN-lt_LT,lv_LV-es_XX,es_XX-lv_LV,lv_LV-hi_IN,hi_IN-lv_LV,lv_LV-zh_CN,zh_CN-lv_LV,my_MM-es_XX,es_XX-my_MM,my_MM-hi_IN,hi_IN-my_MM,my_MM-zh_CN,zh_CN-my_MM,ne_NP-es_XX,es_XX-ne_NP,ne_NP-hi_IN,hi_IN-ne_NP,ne_NP-zh_CN,zh_CN-ne_NP,nl_XX-es_XX,es_XX-nl_XX,nl_XX-hi_IN,hi_IN-nl_XX,nl_XX-zh_CN,zh_CN-nl_XX,ro_RO-es_XX,es_XX-ro_RO,ro_RO-hi_IN,hi_IN-ro_RO,ro_RO-zh_CN,zh_CN-ro_RO,ru_RU-es_XX,es_XX-ru_RU,ru_RU-hi_IN,hi_IN-ru_RU,ru_RU-zh_CN,zh_CN-ru_RU,si_LK-es_XX,es_XX-si_LK,si_LK-hi_IN,hi_IN-si_LK,si_LK-zh_CN,zh_CN-si_LK,tr_TR-es_XX,es_XX-tr_TR,tr_TR-hi_IN,hi_IN-tr_TR,tr_TR-zh_CN,zh_CN-tr_TR,vi_VN-es_XX,es_XX-vi_VN,vi_VN-hi_IN,hi_IN-vi_VN,vi_VN-zh_CN,zh_CN-vi_VN' \ - --lang-dict ${LANG_DICT} --lang-tok-style 'mbart' --sampling-method 'temperature' --sampling-temperature '1.0' > $GEN_TMP_DIR/${SRC}_${TGT}.gen -cat $GEN_TMP_DIR/${SRC}_${TGT}.gen | grep -P "^T-" | cut -f2 | $REPLACE_UNICODE_PUNCT | $NORM_PUNC -l ${TGT:0:2} | $REM_NON_PRINT_CHAR | $TOKENIZER -no-escape ${TGT:0:2} > $GEN_TMP_DIR/${SRC}_${TGT}.hyp -cat $GEN_TMP_DIR/${SRC}_${TGT}.gen | grep -P "^H-" | cut -f3 | $REPLACE_UNICODE_PUNCT | $NORM_PUNC -l ${TGT:0:2} | $REM_NON_PRINT_CHAR | $TOKENIZER -no-escape ${TGT:0:2} > $GEN_TMP_DIR/${SRC}_${TGT}.ref -${MULTIBLEU} $GEN_TMP_DIR/${SRC}_${TGT}.ref < $GEN_TMP_DIR/${SRC}_${TGT}.hyp diff --git a/fairseq/examples/cross_lingual_language_model/README.md b/fairseq/examples/cross_lingual_language_model/README.md deleted file mode 100644 index af9128e39e5925e9411d162c2f24a19e4532d618..0000000000000000000000000000000000000000 --- a/fairseq/examples/cross_lingual_language_model/README.md +++ /dev/null @@ -1,77 +0,0 @@ -# Cross-Lingual Language Model Pre-training - -Below are some details for training Cross-Lingual Language Models (XLM) - similar to the ones presented in [Lample & Conneau, 2019](https://arxiv.org/pdf/1901.07291.pdf) - in Fairseq. The current implementation only supports the Masked Language Model (MLM) from the paper above. - -## Downloading and Tokenizing Monolingual Data - -Pointers to the monolingual data from wikipedia, used for training the XLM-style MLM model as well as details on processing (tokenization and BPE) it can be found in the [XLM Github Repository](https://github.com/facebookresearch/XLM#download--preprocess-monolingual-data). - -Let's assume the following for the code snippets in later sections to work -- Processed data is in the folder: monolingual_data/processed -- Each language has 3 files for train, test and validation. For example we have the following files for English: - train.en, valid.en -- We are training a model for 5 languages: Arabic (ar), German (de), English (en), Hindi (hi) and French (fr) -- The vocabulary file is monolingual_data/processed/vocab_mlm - - -## Fairseq Pre-processing and Binarization - -Pre-process and binarize the data with the MaskedLMDictionary and cross_lingual_lm task - -```bash -# Ensure the output directory exists -DATA_DIR=monolingual_data/fairseq_processed -mkdir -p "$DATA_DIR" - -for lg in ar de en hi fr -do - - fairseq-preprocess \ - --task cross_lingual_lm \ - --srcdict monolingual_data/processed/vocab_mlm \ - --only-source \ - --trainpref monolingual_data/processed/train \ - --validpref monolingual_data/processed/valid \ - --testpref monolingual_data/processed/test \ - --destdir monolingual_data/fairseq_processed \ - --workers 20 \ - --source-lang $lg - - # Since we only have a source language, the output file has a None for the - # target language. Remove this - - for stage in train test valid - - sudo mv "$DATA_DIR/$stage.$lg-None.$lg.bin" "$stage.$lg.bin" - sudo mv "$DATA_DIR/$stage.$lg-None.$lg.idx" "$stage.$lg.idx" - - done - -done -``` - -## Train a Cross-lingual Language Model similar to the XLM MLM model - -Use the following command to train the model on 5 languages. - -``` -fairseq-train \ ---task cross_lingual_lm monolingual_data/fairseq_processed \ ---save-dir checkpoints/mlm \ ---max-update 2400000 --save-interval 1 --no-epoch-checkpoints \ ---arch xlm_base \ ---optimizer adam --lr-scheduler reduce_lr_on_plateau \ ---lr-shrink 0.5 --lr 0.0001 --stop-min-lr 1e-09 \ ---dropout 0.1 \ ---criterion legacy_masked_lm_loss \ ---max-tokens 2048 --tokens-per-sample 256 --attention-dropout 0.1 \ ---dataset-impl lazy --seed 0 \ ---masked-lm-only \ ---monolingual-langs 'ar,de,en,hi,fr' --num-segment 5 \ ---ddp-backend=legacy_ddp -``` - -Some Notes: -- Using tokens_per_sample greater than 256 can cause OOM (out-of-memory) issues. Usually since MLM packs in streams of text, this parameter doesn't need much tuning. -- The Evaluation workflow for computing MLM Perplexity on test data is in progress. -- Finetuning this model on a downstream task is something which is not currently available. diff --git a/fairseq/examples/discriminative_reranking_nmt/README.md b/fairseq/examples/discriminative_reranking_nmt/README.md deleted file mode 100644 index b155e855f2f94e30ad22262f260008fda8ac1804..0000000000000000000000000000000000000000 --- a/fairseq/examples/discriminative_reranking_nmt/README.md +++ /dev/null @@ -1,202 +0,0 @@ -# Discriminative Reranking for Neural Machine Translation -https://aclanthology.org/2021.acl-long.563/ - -This folder contains source code for training DrNMT, a discriminatively trained reranker for neural machine translation. - -## Data preparation -1. Follow the instructions under `examples/translation` to build a base MT model. Prepare three files, one with source sentences, one with ground truth target sentences, and one with hypotheses generated from the base MT model. Each line in the file contains one sentence in raw text (i.e. no sentencepiece, etc.). Below is an example of the files with _N_ hypotheses for each source sentence. - -``` -# Example of the source sentence file: (The file should contain L lines.) - -source_sentence_1 -source_sentence_2 -source_sentence_3 -... -source_sentence_L - -# Example of the target sentence file: (The file should contain L lines.) - -target_sentence_1 -target_sentence_2 -target_sentence_3 -... -target_sentence_L - -# Example of the hypotheses file: (The file should contain L*N lines.) - -source_sentence_1_hypo_1 -source_sentence_1_hypo_2 -... -source_sentence_1_hypo_N -source_sentence_2_hypo_1 -... -source_sentence_2_hypo_N -... -source_sentence_L_hypo_1 -... -source_sentence_L_hypo_N -``` - -2. Download the [XLMR model](https://github.com/fairinternal/fairseq-py/tree/main/examples/xlmr#pre-trained-models). -``` -wget https://dl.fbaipublicfiles.com/fairseq/models/xlmr.base.tar.gz -tar zxvf xlmr.base.tar.gz - -# The folder should contain dict.txt, model.pt and sentencepiece.bpe.model. -``` - -3. Prepare scores and BPE data. -* `N`: Number of hypotheses per each source sentence. We use 50 in the paper. -* `SPLIT`: Name of the data split, i.e. train, valid, test. Use split_name, split_name1, split_name2, ..., if there are multiple datasets for a split, e.g. train, train1, valid, valid1. -* `NUM_SHARDS`: Number of shards. Set this to 1 for non-train splits. -* `METRIC`: The metric for DrNMT to optimize for. We support either `bleu` or `ter`. -``` -# For each data split, e.g. train, valid, test, etc., run the following: - -SOURCE_FILE=/path/to/source_sentence_file -TARGET_FILE=/path/to/target_sentence_file -HYPO_FILE=/path/to/hypo_file -XLMR_DIR=/path/to/xlmr -OUTPUT_DIR=/path/to/output - -python scripts/prep_data.py \ - --input-source ${SOURCE_FILE} \ - --input-target ${TARGET_FILE} \ - --input-hypo ${HYPO_FILE} \ - --output-dir ${OUTPUT_DIR} \ - --split $SPLIT - --beam $N \ - --sentencepiece-model ${XLMR_DIR}/sentencepiece.bpe.model \ - --metric $METRIC \ - --num-shards ${NUM_SHARDS} - -# The script will create ${OUTPUT_DIR}/$METRIC with ${NUM_SHARDS} splits. -# Under split*/input_src, split*/input_tgt and split*/$METRIC, there will be $SPLIT.bpe and $SPLIT.$METRIC files, respectively. - -``` - -4. Pre-process the data into fairseq format. -``` -# use comma to separate if there are more than one train or valid set -for suffix in src tgt ; do - fairseq-preprocess --only-source \ - --trainpref ${OUTPUT_DIR}/$METRIC/split1/input_${suffix}/train.bpe \ - --validpref ${OUTPUT_DIR}/$METRIC/split1/input_${suffix}/valid.bpe \ - --destdir ${OUTPUT_DIR}/$METRIC/split1/input_${suffix} \ - --workers 60 \ - --srcdict ${XLMR_DIR}/dict.txt -done - -for i in `seq 2 ${NUM_SHARDS}`; do - for suffix in src tgt ; do - fairseq-preprocess --only-source \ - --trainpref ${OUTPUT_DIR}/$METRIC/split${i}/input_${suffix}/train.bpe \ - --destdir ${OUTPUT_DIR}/$METRIC/split${i}/input_${suffix} \ - --workers 60 \ - --srcdict ${XLMR_DIR}/dict.txt - - ln -s ${OUTPUT_DIR}/$METRIC/split1/input_${suffix}/valid* ${OUTPUT_DIR}/$METRIC/split${i}/input_${suffix}/. - done - - ln -s ${OUTPUT_DIR}/$METRIC/split1/$METRIC/valid* ${OUTPUT_DIR}/$METRIC/split${i}/$METRIC/. -done -``` - -## Training - -``` -EXP_DIR=/path/to/exp - -# An example of training the model with the config for De-En experiment in the paper. -# The config uses 16 GPUs and 50 hypotheses. -# For training with fewer number of GPUs, set -# distributed_training.distributed_world_size=k +optimization.update_freq='[x]' where x = 16/k -# For training with fewer number of hypotheses, set -# task.mt_beam=N dataset.batch_size=N dataset.required_batch_size_multiple=N - -fairseq-hydra-train -m \ - --config-dir config/ --config-name deen \ - task.data=${OUTPUT_DIR}/$METRIC/split1/ \ - task.num_data_splits=${NUM_SHARDS} \ - model.pretrained_model=${XLMR_DIR}/model.pt \ - common.user_dir=${FAIRSEQ_ROOT}/examples/discriminative_reranking_nmt \ - checkpoint.save_dir=${EXP_DIR} - -``` - -## Inference & scoring -Perform DrNMT reranking (fw + reranker score) -1. Tune weights on valid sets. -``` -# genrate N hypotheses with the base MT model (fw score) -VALID_SOURCE_FILE=/path/to/source_sentences # one sentence per line, converted to the sentencepiece used by the base MT model -VALID_TARGET_FILE=/path/to/target_sentences # one sentence per line in raw text, i.e. no sentencepiece and tokenization -MT_MODEL=/path/to/mt_model -MT_DATA_PATH=/path/to/mt_data - -cat ${VALID_SOURCE_FILE} | \ - fairseq-interactive ${MT_DATA_PATH} \ - --max-tokens 4000 --buffer-size 16 \ - --num-workers 32 --path ${MT_MODEL} \ - --beam $N --nbest $N \ - --post-process sentencepiece &> valid-hypo.out - -# replace "bleu" with "ter" to optimize for TER -python drnmt_rerank.py \ - ${OUTPUT_DIR}/$METRIC/split1/ \ - --path ${EXP_DIR}/checkpoint_best.pt \ - --in-text valid-hypo.out \ - --results-path ${EXP_DIR} \ - --gen-subset valid \ - --target-text ${VALID_TARGET_FILE} \ - --user-dir ${FAIRSEQ_ROOT}/examples/discriminative_reranking_nmt \ - --bpe sentencepiece \ - --sentencepiece-model ${XLMR_DIR}/sentencepiece.bpe.model \ - --beam $N \ - --batch-size $N \ - --metric bleu \ - --tune - -``` - -2. Apply best weights on test sets -``` -# genrate N hypotheses with the base MT model (fw score) -TEST_SOURCE_FILE=/path/to/source_sentences # one sentence per line, converted to the sentencepiece used by the base MT model - -cat ${TEST_SOURCE_FILE} | \ - fairseq-interactive ${MT_DATA_PATH} \ - --max-tokens 4000 --buffer-size 16 \ - --num-workers 32 --path ${MT_MODEL} \ - --beam $N --nbest $N \ - --post-process sentencepiece &> test-hypo.out - -# replace "bleu" with "ter" to evaluate TER -# Add --target-text for evaluating BLEU/TER, -# otherwise the script will only generate the hypotheses with the highest scores only. -python drnmt_rerank.py \ - ${OUTPUT_DIR}/$METRIC/split1/ \ - --path ${EXP_DIR}/checkpoint_best.pt \ - --in-text test-hypo.out \ - --results-path ${EXP_DIR} \ - --gen-subset test \ - --user-dir ${FAIRSEQ_ROOT}/examples/discriminative_reranking_nmt \ - --bpe sentencepiece \ - --sentencepiece-model ${XLMR_DIR}/sentencepiece.bpe.model \ - --beam $N \ - --batch-size $N \ - --metric bleu \ - --fw-weight ${BEST_FW_WEIGHT} \ - --lenpen ${BEST_LENPEN} -``` - -## Citation -```bibtex -@inproceedings{lee2021discriminative, - title={Discriminative Reranking for Neural Machine Translation}, - author={Lee, Ann and Auli, Michael and Ranzato, Marc'Aurelio}, - booktitle={ACL}, - year={2021} -} -``` diff --git a/fairseq/examples/discriminative_reranking_nmt/__init__.py b/fairseq/examples/discriminative_reranking_nmt/__init__.py deleted file mode 100644 index 0278f6a27340c7ff7e207d09348483d1b0d3a100..0000000000000000000000000000000000000000 --- a/fairseq/examples/discriminative_reranking_nmt/__init__.py +++ /dev/null @@ -1 +0,0 @@ -from . import criterions, models, tasks # noqa diff --git a/fairseq/examples/discriminative_reranking_nmt/config/deen.yaml b/fairseq/examples/discriminative_reranking_nmt/config/deen.yaml deleted file mode 100644 index 3fc2d5fcf5bacbb842d181fcfcde80e55331fed7..0000000000000000000000000000000000000000 --- a/fairseq/examples/discriminative_reranking_nmt/config/deen.yaml +++ /dev/null @@ -1,56 +0,0 @@ -# @package _group_ - -common: - fp16: true - log_format: json - log_interval: 50 - seed: 2 - -checkpoint: - no_epoch_checkpoints: true - best_checkpoint_metric: bleu - maximize_best_checkpoint_metric: true - -task: - _name: discriminative_reranking_nmt - data: ??? - num_data_splits: ??? - include_src: true - mt_beam: 50 - eval_target_metric: true - target_metric: bleu - -dataset: - batch_size: 50 - num_workers: 6 - required_batch_size_multiple: 50 - valid_subset: ??? - -criterion: - _name: kl_divergence_rereanking - target_dist_norm: minmax - temperature: 0.5 - -optimization: - max_epoch: 200 - lr: [0.00005] - update_freq: [32] - -optimizer: - _name: adam - adam_betas: (0.9,0.98) - adam_eps: 1e-06 - -lr_scheduler: - _name: polynomial_decay - warmup_updates: 8000 - total_num_update: 320000 - -model: - _name: discriminative_nmt_reranker - pretrained_model: ??? - classifier_dropout: 0.2 - -distributed_training: - ddp_backend: no_c10d - distributed_world_size: 16 diff --git a/fairseq/examples/discriminative_reranking_nmt/criterions/__init__.py b/fairseq/examples/discriminative_reranking_nmt/criterions/__init__.py deleted file mode 100644 index 7c257c2700f015cb123a976584aef72f0429eb0c..0000000000000000000000000000000000000000 --- a/fairseq/examples/discriminative_reranking_nmt/criterions/__init__.py +++ /dev/null @@ -1,6 +0,0 @@ -from .discriminative_reranking_criterion import KLDivergenceRerankingCriterion - - -__all__ = [ - "KLDivergenceRerankingCriterion", -] diff --git a/fairseq/examples/discriminative_reranking_nmt/criterions/discriminative_reranking_criterion.py b/fairseq/examples/discriminative_reranking_nmt/criterions/discriminative_reranking_criterion.py deleted file mode 100644 index 0b02ce18772454697e61f827d96d76ad361b9cd1..0000000000000000000000000000000000000000 --- a/fairseq/examples/discriminative_reranking_nmt/criterions/discriminative_reranking_criterion.py +++ /dev/null @@ -1,138 +0,0 @@ -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. - -import math -from dataclasses import dataclass, field - -import torch -import torch.nn.functional as F - -from fairseq import metrics, utils -from fairseq.criterions import FairseqCriterion, register_criterion -from fairseq.dataclass import ChoiceEnum, FairseqDataclass - - -_EPSILON = torch.finfo(torch.float32).eps -TARGET_DIST_NORM_CHOICES = ChoiceEnum(["none", "minmax"]) - - -@dataclass -class KLDivergenceRerankingCriterionConfig(FairseqDataclass): - target_dist_norm: TARGET_DIST_NORM_CHOICES = field( - default="none", - metadata={"help": "method to normalize the range of target scores"}, - ) - temperature: float = field( - default=1.0, - metadata={"help": "temperature in softmax for target distributions"}, - ) - forward_batch_size: int = field( - default=32, - metadata={ - "help": "number of hypotheses per batch for model forward (set a value smaller than --mt-beam to avoid OOM when training with a large beam size)" - }, - ) - - -@register_criterion( - "kl_divergence_rereanking", dataclass=KLDivergenceRerankingCriterionConfig -) -class KLDivergenceRerankingCriterion(FairseqCriterion): - def __init__( - self, task, target_dist_norm, temperature, forward_batch_size, - ): - super().__init__(task) - self.target_dist_norm = target_dist_norm - self.temperature = temperature - self.forward_batch_size = forward_batch_size - - def forward(self, model, sample, reduce=True): - """Compute the loss for the given sample. - - Returns a tuple with three elements: - 1) the loss - 2) the sample size, which is used as the denominator for the gradient - 3) logging outputs to display while training - """ - - sample_size = sample["id"].numel() - assert sample_size % self.task.cfg.mt_beam == 0, ( - f"sample_size ({sample_size}) cannot be divided by beam size ({self.task.cfg.mt_beam})." - f"Please set --required-batch-size-multiple={self.task.cfg.mt_beam}." - ) - - # split into smaller batches for model forward - batch_out = [] - for i in range(0, sample_size, self.forward_batch_size): - j = min(i + self.forward_batch_size, sample_size) - - out = model( - src_tokens=sample["net_input"]["src_tokens"][i:j, :], - src_lengths=sample["net_input"]["src_lengths"][i:j], - ) - - batch_out.append( - model.sentence_forward(out, sample["net_input"]["src_tokens"][i:j, :]) - ) - - batch_out = torch.cat(batch_out, dim=0).view( - self.task.cfg.mt_beam, sample_size // self.task.cfg.mt_beam, -1 - ) # T x B x C - if model.joint_classification == "sent": - batch_out = model.joint_forward(batch_out) - scores = model.classification_forward(batch_out.view(sample_size, 1, -1)).view( - -1, self.task.cfg.mt_beam - ) # input: B x T x C - - loss = self.compute_kl_loss( - scores, sample["target"][:, 0].view(-1, self.task.cfg.mt_beam) - ) - - sample_size = sample_size // self.task.cfg.mt_beam - - logging_output = { - "loss": loss.detach(), - "ntokens": sample["ntokens"], - "nsentences": sample_size * self.task.cfg.mt_beam, - "sample_size": sample_size, - "scores": scores.detach(), - } - - return loss, sample_size, logging_output - - def compute_kl_loss(self, logits, target): - norm_target = target - if self.target_dist_norm == "minmax": - min_v = torch.min(target, 1, keepdim=True).values - max_v = torch.max(target, 1, keepdim=True).values - norm_target = (target - min_v) / (max_v - min_v + _EPSILON) - - target_dist = F.softmax( - norm_target / self.temperature, dim=-1, dtype=torch.float32 - ) - model_dist = F.log_softmax(logits, dim=-1, dtype=torch.float32) - loss = -(target_dist * model_dist - target_dist * target_dist.log()).sum() - return loss - - @staticmethod - def reduce_metrics(logging_outputs) -> None: - """Aggregate logging outputs from data parallel training.""" - loss_sum = utils.item(sum(log.get("loss", 0) for log in logging_outputs)) - - sample_size = utils.item( - sum(log.get("sample_size", 0) for log in logging_outputs) - ) - - loss = loss_sum / sample_size / math.log(2) - metrics.log_scalar("loss", loss, sample_size, round=3) - - @staticmethod - def logging_outputs_can_be_summed() -> bool: - """ - Whether the logging outputs returned by `forward` can be summed - across workers prior to calling `reduce_metrics`. Setting this - to True will improves distributed training speed. - """ - return True diff --git a/fairseq/examples/discriminative_reranking_nmt/drnmt_rerank.py b/fairseq/examples/discriminative_reranking_nmt/drnmt_rerank.py deleted file mode 100644 index 2e0fc2bd29aedb0b477b7cc8e2c3b606acdd454a..0000000000000000000000000000000000000000 --- a/fairseq/examples/discriminative_reranking_nmt/drnmt_rerank.py +++ /dev/null @@ -1,364 +0,0 @@ -#!/usr/bin/env python3 -u -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. -""" -Score raw text with a trained model. -""" - -from collections import namedtuple -import logging -from multiprocessing import Pool -import sys -import os -import random - -import numpy as np -import sacrebleu -import torch - -from fairseq import checkpoint_utils, options, utils - - -logger = logging.getLogger("fairseq_cli.drnmt_rerank") -logger.setLevel(logging.INFO) - -Batch = namedtuple("Batch", "ids src_tokens src_lengths") - - -pool_init_variables = {} - - -def init_loaded_scores(mt_scores, model_scores, hyp, ref): - global pool_init_variables - pool_init_variables["mt_scores"] = mt_scores - pool_init_variables["model_scores"] = model_scores - pool_init_variables["hyp"] = hyp - pool_init_variables["ref"] = ref - - -def parse_fairseq_gen(filename, task): - source = {} - hypos = {} - scores = {} - with open(filename, "r", encoding="utf-8") as f: - for line in f: - line = line.strip() - if line.startswith("S-"): # source - uid, text = line.split("\t", 1) - uid = int(uid[2:]) - source[uid] = text - elif line.startswith("D-"): # hypo - uid, score, text = line.split("\t", 2) - uid = int(uid[2:]) - if uid not in hypos: - hypos[uid] = [] - scores[uid] = [] - hypos[uid].append(text) - scores[uid].append(float(score)) - else: - continue - - source_out = [source[i] for i in range(len(hypos))] - hypos_out = [h for i in range(len(hypos)) for h in hypos[i]] - scores_out = [s for i in range(len(scores)) for s in scores[i]] - - return source_out, hypos_out, scores_out - - -def read_target(filename): - with open(filename, "r", encoding="utf-8") as f: - output = [line.strip() for line in f] - return output - - -def make_batches(args, src, hyp, task, max_positions, encode_fn): - assert len(src) * args.beam == len( - hyp - ), f"Expect {len(src) * args.beam} hypotheses for {len(src)} source sentences with beam size {args.beam}. Got {len(hyp)} hypotheses intead." - hyp_encode = [ - task.source_dictionary.encode_line(encode_fn(h), add_if_not_exist=False).long() - for h in hyp - ] - if task.cfg.include_src: - src_encode = [ - task.source_dictionary.encode_line( - encode_fn(s), add_if_not_exist=False - ).long() - for s in src - ] - tokens = [(src_encode[i // args.beam], h) for i, h in enumerate(hyp_encode)] - lengths = [(t1.numel(), t2.numel()) for t1, t2 in tokens] - else: - tokens = [(h,) for h in hyp_encode] - lengths = [(h.numel(),) for h in hyp_encode] - - itr = task.get_batch_iterator( - dataset=task.build_dataset_for_inference(tokens, lengths), - max_tokens=args.max_tokens, - max_sentences=args.batch_size, - max_positions=max_positions, - ignore_invalid_inputs=args.skip_invalid_size_inputs_valid_test, - ).next_epoch_itr(shuffle=False) - - for batch in itr: - yield Batch( - ids=batch["id"], - src_tokens=batch["net_input"]["src_tokens"], - src_lengths=batch["net_input"]["src_lengths"], - ) - - -def decode_rerank_scores(args): - if args.max_tokens is None and args.batch_size is None: - args.batch_size = 1 - - logger.info(args) - - use_cuda = torch.cuda.is_available() and not args.cpu - - # Load ensemble - logger.info("loading model(s) from {}".format(args.path)) - models, _model_args, task = checkpoint_utils.load_model_ensemble_and_task( - [args.path], arg_overrides=eval(args.model_overrides), - ) - - for model in models: - if args.fp16: - model.half() - if use_cuda: - model.cuda() - - # Initialize generator - generator = task.build_generator(args) - - # Handle tokenization and BPE - tokenizer = task.build_tokenizer(args) - bpe = task.build_bpe(args) - - def encode_fn(x): - if tokenizer is not None: - x = tokenizer.encode(x) - if bpe is not None: - x = bpe.encode(x) - return x - - max_positions = utils.resolve_max_positions( - task.max_positions(), *[model.max_positions() for model in models] - ) - - src, hyp, mt_scores = parse_fairseq_gen(args.in_text, task) - model_scores = {} - logger.info("decode reranker score") - for batch in make_batches(args, src, hyp, task, max_positions, encode_fn): - src_tokens = batch.src_tokens - src_lengths = batch.src_lengths - if use_cuda: - src_tokens = src_tokens.cuda() - src_lengths = src_lengths.cuda() - - sample = { - "net_input": {"src_tokens": src_tokens, "src_lengths": src_lengths}, - } - scores = task.inference_step(generator, models, sample) - - for id, sc in zip(batch.ids.tolist(), scores.tolist()): - model_scores[id] = sc[0] - - model_scores = [model_scores[i] for i in range(len(model_scores))] - - return src, hyp, mt_scores, model_scores - - -def get_score(mt_s, md_s, w1, lp, tgt_len): - return mt_s / (tgt_len ** lp) * w1 + md_s - - -def get_best_hyps(mt_scores, md_scores, hypos, fw_weight, lenpen, beam): - assert len(mt_scores) == len(md_scores) and len(mt_scores) == len(hypos) - hypo_scores = [] - best_hypos = [] - best_scores = [] - offset = 0 - for i in range(len(hypos)): - tgt_len = len(hypos[i].split()) - hypo_scores.append( - get_score(mt_scores[i], md_scores[i], fw_weight, lenpen, tgt_len) - ) - - if (i + 1) % beam == 0: - max_i = np.argmax(hypo_scores) - best_hypos.append(hypos[offset + max_i]) - best_scores.append(hypo_scores[max_i]) - hypo_scores = [] - offset += beam - return best_hypos, best_scores - - -def eval_metric(args, hypos, ref): - if args.metric == "bleu": - score = sacrebleu.corpus_bleu(hypos, [ref]).score - else: - score = sacrebleu.corpus_ter(hypos, [ref]).score - - return score - - -def score_target_hypo(args, fw_weight, lp): - mt_scores = pool_init_variables["mt_scores"] - model_scores = pool_init_variables["model_scores"] - hyp = pool_init_variables["hyp"] - ref = pool_init_variables["ref"] - best_hypos, _ = get_best_hyps( - mt_scores, model_scores, hyp, fw_weight, lp, args.beam - ) - rerank_eval = None - if ref: - rerank_eval = eval_metric(args, best_hypos, ref) - print(f"fw_weight {fw_weight}, lenpen {lp}, eval {rerank_eval}") - - return rerank_eval - - -def print_result(best_scores, best_hypos, output_file): - for i, (s, h) in enumerate(zip(best_scores, best_hypos)): - print(f"{i}\t{s}\t{h}", file=output_file) - - -def main(args): - utils.import_user_module(args) - - src, hyp, mt_scores, model_scores = decode_rerank_scores(args) - - assert ( - not args.tune or args.target_text is not None - ), "--target-text has to be set when tuning weights" - if args.target_text: - ref = read_target(args.target_text) - assert len(src) == len( - ref - ), f"different numbers of source and target sentences ({len(src)} vs. {len(ref)})" - - orig_best_hypos = [hyp[i] for i in range(0, len(hyp), args.beam)] - orig_eval = eval_metric(args, orig_best_hypos, ref) - - if args.tune: - logger.info("tune weights for reranking") - - random_params = np.array( - [ - [ - random.uniform( - args.lower_bound_fw_weight, args.upper_bound_fw_weight - ), - random.uniform(args.lower_bound_lenpen, args.upper_bound_lenpen), - ] - for k in range(args.num_trials) - ] - ) - - logger.info("launching pool") - with Pool( - 32, - initializer=init_loaded_scores, - initargs=(mt_scores, model_scores, hyp, ref), - ) as p: - rerank_scores = p.starmap( - score_target_hypo, - [ - (args, random_params[i][0], random_params[i][1],) - for i in range(args.num_trials) - ], - ) - if args.metric == "bleu": - best_index = np.argmax(rerank_scores) - else: - best_index = np.argmin(rerank_scores) - best_fw_weight = random_params[best_index][0] - best_lenpen = random_params[best_index][1] - else: - assert ( - args.lenpen is not None and args.fw_weight is not None - ), "--lenpen and --fw-weight should be set" - best_fw_weight, best_lenpen = args.fw_weight, args.lenpen - - best_hypos, best_scores = get_best_hyps( - mt_scores, model_scores, hyp, best_fw_weight, best_lenpen, args.beam - ) - - if args.results_path is not None: - os.makedirs(args.results_path, exist_ok=True) - output_path = os.path.join( - args.results_path, "generate-{}.txt".format(args.gen_subset), - ) - with open(output_path, "w", buffering=1, encoding="utf-8") as o: - print_result(best_scores, best_hypos, o) - else: - print_result(best_scores, best_hypos, sys.stdout) - - if args.target_text: - rerank_eval = eval_metric(args, best_hypos, ref) - print(f"before reranking, {args.metric.upper()}:", orig_eval) - print( - f"after reranking with fw_weight={best_fw_weight}, lenpen={best_lenpen}, {args.metric.upper()}:", - rerank_eval, - ) - - -def cli_main(): - parser = options.get_generation_parser(interactive=True) - - parser.add_argument( - "--in-text", - default=None, - required=True, - help="text from fairseq-interactive output, containing source sentences and hypotheses", - ) - parser.add_argument("--target-text", default=None, help="reference text") - parser.add_argument("--metric", type=str, choices=["bleu", "ter"], default="bleu") - parser.add_argument( - "--tune", - action="store_true", - help="if set, tune weights on fw scores and lenpen instead of applying fixed weights for reranking", - ) - parser.add_argument( - "--lower-bound-fw-weight", - default=0.0, - type=float, - help="lower bound of search space", - ) - parser.add_argument( - "--upper-bound-fw-weight", - default=3, - type=float, - help="upper bound of search space", - ) - parser.add_argument( - "--lower-bound-lenpen", - default=0.0, - type=float, - help="lower bound of search space", - ) - parser.add_argument( - "--upper-bound-lenpen", - default=3, - type=float, - help="upper bound of search space", - ) - parser.add_argument( - "--fw-weight", type=float, default=None, help="weight on the fw model score" - ) - parser.add_argument( - "--num-trials", - default=1000, - type=int, - help="number of trials to do for random search", - ) - - args = options.parse_args_and_arch(parser) - main(args) - - -if __name__ == "__main__": - cli_main() diff --git a/fairseq/examples/discriminative_reranking_nmt/models/__init__.py b/fairseq/examples/discriminative_reranking_nmt/models/__init__.py deleted file mode 100644 index c593ea5f1842794bfcc952fc93c679a5f16aeb98..0000000000000000000000000000000000000000 --- a/fairseq/examples/discriminative_reranking_nmt/models/__init__.py +++ /dev/null @@ -1,6 +0,0 @@ -from .discriminative_reranking_model import DiscriminativeNMTReranker - - -__all__ = [ - "DiscriminativeNMTReranker", -] diff --git a/fairseq/examples/discriminative_reranking_nmt/models/discriminative_reranking_model.py b/fairseq/examples/discriminative_reranking_nmt/models/discriminative_reranking_model.py deleted file mode 100644 index e4b5887f825df36f4e1e0384f38fefe790e485e6..0000000000000000000000000000000000000000 --- a/fairseq/examples/discriminative_reranking_nmt/models/discriminative_reranking_model.py +++ /dev/null @@ -1,365 +0,0 @@ -from dataclasses import dataclass, field -import os - -import torch -import torch.nn as nn - -from fairseq import utils -from fairseq.dataclass import ChoiceEnum, FairseqDataclass -from fairseq.models import ( - BaseFairseqModel, - register_model, -) - -from fairseq.models.roberta.model import RobertaClassificationHead - -from fairseq.modules import ( - LayerNorm, - TransformerSentenceEncoder, - TransformerSentenceEncoderLayer, -) - - -ACTIVATION_FN_CHOICES = ChoiceEnum(utils.get_available_activation_fns()) -JOINT_CLASSIFICATION_CHOICES = ChoiceEnum(["none", "sent"]) -SENTENCE_REP_CHOICES = ChoiceEnum(["head", "meanpool", "maxpool"]) - - -def update_init_roberta_model_state(state): - """ - update the state_dict of a Roberta model for initializing - weights of the BertRanker - """ - for k in list(state.keys()): - if ".lm_head." in k or "version" in k: - del state[k] - continue - # remove 'encoder/decoder.sentence_encoder.' from the key - assert k.startswith("encoder.sentence_encoder.") or k.startswith( - "decoder.sentence_encoder." - ), f"Cannot recognize parameter name {k}" - if "layernorm_embedding" in k: - new_k = k.replace(".layernorm_embedding.", ".emb_layer_norm.") - state[new_k[25:]] = state[k] - else: - state[k[25:]] = state[k] - del state[k] - - -class BaseRanker(nn.Module): - def __init__(self, args, task): - super().__init__() - - self.separator_token = task.dictionary.eos() - self.padding_idx = task.dictionary.pad() - - def forward(self, src_tokens): - raise NotImplementedError - - def get_segment_labels(self, src_tokens): - segment_boundary = (src_tokens == self.separator_token).long() - segment_labels = ( - segment_boundary.cumsum(dim=1) - - segment_boundary - - (src_tokens == self.padding_idx).long() - ) - - return segment_labels - - def get_positions(self, src_tokens, segment_labels): - segment_positions = ( - torch.arange(src_tokens.shape[1]) - .to(src_tokens.device) - .repeat(src_tokens.shape[0], 1) - ) - segment_boundary = (src_tokens == self.separator_token).long() - _, col_idx = (segment_positions * segment_boundary).nonzero(as_tuple=True) - col_idx = torch.cat([torch.zeros(1).type_as(col_idx), col_idx]) - offset = torch.cat( - [ - torch.zeros(1).type_as(segment_boundary), - segment_boundary.sum(dim=1).cumsum(dim=0)[:-1], - ] - ) - segment_positions -= col_idx[segment_labels + offset.unsqueeze(1)] * ( - segment_labels != 0 - ) - - padding_mask = src_tokens.ne(self.padding_idx) - segment_positions = (segment_positions + 1) * padding_mask.type_as( - segment_positions - ) + self.padding_idx - - return segment_positions - - -class BertRanker(BaseRanker): - def __init__(self, args, task): - super(BertRanker, self).__init__(args, task) - - init_model = getattr(args, "pretrained_model", "") - self.joint_layers = nn.ModuleList() - if os.path.isfile(init_model): - print(f"initialize weight from {init_model}") - - from fairseq import hub_utils - - x = hub_utils.from_pretrained( - os.path.dirname(init_model), - checkpoint_file=os.path.basename(init_model), - ) - - in_state_dict = x["models"][0].state_dict() - init_args = x["args"].model - - num_positional_emb = init_args.max_positions + task.dictionary.pad() + 1 - - # follow the setup in roberta - self.model = TransformerSentenceEncoder( - padding_idx=task.dictionary.pad(), - vocab_size=len(task.dictionary), - num_encoder_layers=getattr( - args, "encoder_layers", init_args.encoder_layers - ), - embedding_dim=init_args.encoder_embed_dim, - ffn_embedding_dim=init_args.encoder_ffn_embed_dim, - num_attention_heads=init_args.encoder_attention_heads, - dropout=init_args.dropout, - attention_dropout=init_args.attention_dropout, - activation_dropout=init_args.activation_dropout, - num_segments=2, # add language embeddings - max_seq_len=num_positional_emb, - offset_positions_by_padding=False, - encoder_normalize_before=True, - apply_bert_init=True, - activation_fn=init_args.activation_fn, - freeze_embeddings=args.freeze_embeddings, - n_trans_layers_to_freeze=args.n_trans_layers_to_freeze, - ) - - # still need to learn segment embeddings as we added a second language embedding - if args.freeze_embeddings: - for p in self.model.segment_embeddings.parameters(): - p.requires_grad = False - - update_init_roberta_model_state(in_state_dict) - print("loading weights from the pretrained model") - self.model.load_state_dict( - in_state_dict, strict=False - ) # ignore mismatch in language embeddings - - ffn_embedding_dim = init_args.encoder_ffn_embed_dim - num_attention_heads = init_args.encoder_attention_heads - dropout = init_args.dropout - attention_dropout = init_args.attention_dropout - activation_dropout = init_args.activation_dropout - activation_fn = init_args.activation_fn - - classifier_embed_dim = getattr( - args, "embed_dim", init_args.encoder_embed_dim - ) - if classifier_embed_dim != init_args.encoder_embed_dim: - self.transform_layer = nn.Linear( - init_args.encoder_embed_dim, classifier_embed_dim - ) - else: - self.model = TransformerSentenceEncoder( - padding_idx=task.dictionary.pad(), - vocab_size=len(task.dictionary), - num_encoder_layers=args.encoder_layers, - embedding_dim=args.embed_dim, - ffn_embedding_dim=args.ffn_embed_dim, - num_attention_heads=args.attention_heads, - dropout=args.dropout, - attention_dropout=args.attention_dropout, - activation_dropout=args.activation_dropout, - max_seq_len=task.max_positions() - if task.max_positions() - else args.tokens_per_sample, - num_segments=2, - offset_positions_by_padding=False, - encoder_normalize_before=args.encoder_normalize_before, - apply_bert_init=args.apply_bert_init, - activation_fn=args.activation_fn, - ) - - classifier_embed_dim = args.embed_dim - ffn_embedding_dim = args.ffn_embed_dim - num_attention_heads = args.attention_heads - dropout = args.dropout - attention_dropout = args.attention_dropout - activation_dropout = args.activation_dropout - activation_fn = args.activation_fn - - self.joint_classification = args.joint_classification - if args.joint_classification == "sent": - if args.joint_normalize_before: - self.joint_layer_norm = LayerNorm(classifier_embed_dim) - else: - self.joint_layer_norm = None - - self.joint_layers = nn.ModuleList( - [ - TransformerSentenceEncoderLayer( - embedding_dim=classifier_embed_dim, - ffn_embedding_dim=ffn_embedding_dim, - num_attention_heads=num_attention_heads, - dropout=dropout, - attention_dropout=attention_dropout, - activation_dropout=activation_dropout, - activation_fn=activation_fn, - ) - for _ in range(args.num_joint_layers) - ] - ) - - self.classifier = RobertaClassificationHead( - classifier_embed_dim, - classifier_embed_dim, - 1, # num_classes - "tanh", - args.classifier_dropout, - ) - - def forward(self, src_tokens, src_lengths): - segment_labels = self.get_segment_labels(src_tokens) - positions = self.get_positions(src_tokens, segment_labels) - - inner_states, _ = self.model( - tokens=src_tokens, - segment_labels=segment_labels, - last_state_only=True, - positions=positions, - ) - - return inner_states[-1].transpose(0, 1) # T x B x C -> B x T x C - - def sentence_forward(self, encoder_out, src_tokens=None, sentence_rep="head"): - # encoder_out: B x T x C - if sentence_rep == "head": - x = encoder_out[:, :1, :] - else: # 'meanpool', 'maxpool' - assert src_tokens is not None, "meanpool requires src_tokens input" - segment_labels = self.get_segment_labels(src_tokens) - padding_mask = src_tokens.ne(self.padding_idx) - encoder_mask = segment_labels * padding_mask.type_as(segment_labels) - - if sentence_rep == "meanpool": - ntokens = torch.sum(encoder_mask, dim=1, keepdim=True) - x = torch.sum( - encoder_out * encoder_mask.unsqueeze(2), dim=1, keepdim=True - ) / ntokens.unsqueeze(2).type_as(encoder_out) - else: # 'maxpool' - encoder_out[ - (encoder_mask == 0).unsqueeze(2).repeat(1, 1, encoder_out.shape[-1]) - ] = -float("inf") - x, _ = torch.max(encoder_out, dim=1, keepdim=True) - - if hasattr(self, "transform_layer"): - x = self.transform_layer(x) - - return x # B x 1 x C - - def joint_forward(self, x): - # x: T x B x C - if self.joint_layer_norm: - x = self.joint_layer_norm(x.transpose(0, 1)) - x = x.transpose(0, 1) - - for layer in self.joint_layers: - x, _ = layer(x, self_attn_padding_mask=None) - return x - - def classification_forward(self, x): - # x: B x T x C - return self.classifier(x) - - -@dataclass -class DiscriminativeNMTRerankerConfig(FairseqDataclass): - pretrained_model: str = field( - default="", metadata={"help": "pretrained model to load"} - ) - sentence_rep: SENTENCE_REP_CHOICES = field( - default="head", - metadata={ - "help": "method to transform the output of the transformer stack to a sentence-level representation" - }, - ) - - dropout: float = field(default=0.1, metadata={"help": "dropout probability"}) - attention_dropout: float = field( - default=0.0, metadata={"help": "dropout probability for attention weights"} - ) - activation_dropout: float = field( - default=0.0, metadata={"help": "dropout probability after activation in FFN"} - ) - classifier_dropout: float = field( - default=0.0, metadata={"help": "classifier dropout probability"} - ) - embed_dim: int = field(default=768, metadata={"help": "embedding dimension"}) - ffn_embed_dim: int = field( - default=2048, metadata={"help": "embedding dimension for FFN"} - ) - encoder_layers: int = field(default=12, metadata={"help": "num encoder layers"}) - attention_heads: int = field(default=8, metadata={"help": "num attention heads"}) - encoder_normalize_before: bool = field( - default=False, metadata={"help": "apply layernorm before each encoder block"} - ) - apply_bert_init: bool = field( - default=False, metadata={"help": "use custom param initialization for BERT"} - ) - activation_fn: ACTIVATION_FN_CHOICES = field( - default="relu", metadata={"help": "activation function to use"} - ) - freeze_embeddings: bool = field( - default=False, metadata={"help": "freeze embeddings in the pretrained model"} - ) - n_trans_layers_to_freeze: int = field( - default=0, - metadata={ - "help": "number of layers to freeze in the pretrained transformer model" - }, - ) - - # joint classfication - joint_classification: JOINT_CLASSIFICATION_CHOICES = field( - default="none", - metadata={"help": "method to compute joint features for classification"}, - ) - num_joint_layers: int = field( - default=1, metadata={"help": "number of joint layers"} - ) - joint_normalize_before: bool = field( - default=False, - metadata={"help": "apply layer norm on the input to the joint layer"}, - ) - - -@register_model( - "discriminative_nmt_reranker", dataclass=DiscriminativeNMTRerankerConfig -) -class DiscriminativeNMTReranker(BaseFairseqModel): - @classmethod - def build_model(cls, args, task): - model = BertRanker(args, task) - return DiscriminativeNMTReranker(args, model) - - def __init__(self, args, model): - super().__init__() - - self.model = model - self.sentence_rep = args.sentence_rep - self.joint_classification = args.joint_classification - - def forward(self, src_tokens, src_lengths, **kwargs): - return self.model(src_tokens, src_lengths) - - def sentence_forward(self, encoder_out, src_tokens): - return self.model.sentence_forward(encoder_out, src_tokens, self.sentence_rep) - - def joint_forward(self, x): - return self.model.joint_forward(x) - - def classification_forward(self, x): - return self.model.classification_forward(x) diff --git a/fairseq/examples/discriminative_reranking_nmt/scripts/prep_data.py b/fairseq/examples/discriminative_reranking_nmt/scripts/prep_data.py deleted file mode 100755 index 7aa7d37edc2c3e4c1d293911b753abf2ef597a7e..0000000000000000000000000000000000000000 --- a/fairseq/examples/discriminative_reranking_nmt/scripts/prep_data.py +++ /dev/null @@ -1,136 +0,0 @@ -#!/usr/bin/env python - -import argparse -from multiprocessing import Pool -from pathlib import Path - -import sacrebleu -import sentencepiece as spm - - -def read_text_file(filename): - with open(filename, "r") as f: - output = [line.strip() for line in f] - - return output - - -def get_bleu(in_sent, target_sent): - bleu = sacrebleu.corpus_bleu([in_sent], [[target_sent]]) - out = " ".join( - map(str, [bleu.score, bleu.sys_len, bleu.ref_len] + bleu.counts + bleu.totals) - ) - return out - - -def get_ter(in_sent, target_sent): - ter = sacrebleu.corpus_ter([in_sent], [[target_sent]]) - out = " ".join(map(str, [ter.score, ter.num_edits, ter.ref_length])) - return out - - -def init(sp_model): - global sp - sp = spm.SentencePieceProcessor() - sp.Load(sp_model) - - -def process(source_sent, target_sent, hypo_sent, metric): - source_bpe = " ".join(sp.EncodeAsPieces(source_sent)) - hypo_bpe = [" ".join(sp.EncodeAsPieces(h)) for h in hypo_sent] - - if metric == "bleu": - score_str = [get_bleu(h, target_sent) for h in hypo_sent] - else: # ter - score_str = [get_ter(h, target_sent) for h in hypo_sent] - - return source_bpe, hypo_bpe, score_str - - -def main(args): - assert ( - args.split.startswith("train") or args.num_shards == 1 - ), "--num-shards should be set to 1 for valid and test sets" - assert ( - args.split.startswith("train") - or args.split.startswith("valid") - or args.split.startswith("test") - ), "--split should be set to train[n]/valid[n]/test[n]" - - source_sents = read_text_file(args.input_source) - target_sents = read_text_file(args.input_target) - - num_sents = len(source_sents) - assert num_sents == len( - target_sents - ), f"{args.input_source} and {args.input_target} should have the same number of sentences." - - hypo_sents = read_text_file(args.input_hypo) - assert ( - len(hypo_sents) % args.beam == 0 - ), f"Number of hypotheses ({len(hypo_sents)}) cannot be divided by beam size ({args.beam})." - - hypo_sents = [ - hypo_sents[i : i + args.beam] for i in range(0, len(hypo_sents), args.beam) - ] - assert num_sents == len( - hypo_sents - ), f"{args.input_hypo} should contain {num_sents * args.beam} hypotheses but only has {len(hypo_sents) * args.beam}. (--beam={args.beam})" - - output_dir = args.output_dir / args.metric - for ns in range(args.num_shards): - print(f"processing shard {ns+1}/{args.num_shards}") - shard_output_dir = output_dir / f"split{ns+1}" - source_output_dir = shard_output_dir / "input_src" - hypo_output_dir = shard_output_dir / "input_tgt" - metric_output_dir = shard_output_dir / args.metric - - source_output_dir.mkdir(parents=True, exist_ok=True) - hypo_output_dir.mkdir(parents=True, exist_ok=True) - metric_output_dir.mkdir(parents=True, exist_ok=True) - - if args.n_proc > 1: - with Pool( - args.n_proc, initializer=init, initargs=(args.sentencepiece_model,) - ) as p: - output = p.starmap( - process, - [ - (source_sents[i], target_sents[i], hypo_sents[i], args.metric) - for i in range(ns, num_sents, args.num_shards) - ], - ) - else: - init(args.sentencepiece_model) - output = [ - process(source_sents[i], target_sents[i], hypo_sents[i], args.metric) - for i in range(ns, num_sents, args.num_shards) - ] - - with open(source_output_dir / f"{args.split}.bpe", "w") as s_o, open( - hypo_output_dir / f"{args.split}.bpe", "w" - ) as h_o, open(metric_output_dir / f"{args.split}.{args.metric}", "w") as m_o: - for source_bpe, hypo_bpe, score_str in output: - assert len(hypo_bpe) == len(score_str) - for h, m in zip(hypo_bpe, score_str): - s_o.write(f"{source_bpe}\n") - h_o.write(f"{h}\n") - m_o.write(f"{m}\n") - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("--input-source", type=Path, required=True) - parser.add_argument("--input-target", type=Path, required=True) - parser.add_argument("--input-hypo", type=Path, required=True) - parser.add_argument("--output-dir", type=Path, required=True) - parser.add_argument("--split", type=str, required=True) - parser.add_argument("--beam", type=int, required=True) - parser.add_argument("--sentencepiece-model", type=str, required=True) - parser.add_argument("--metric", type=str, choices=["bleu", "ter"], default="bleu") - parser.add_argument("--num-shards", type=int, default=1) - parser.add_argument("--n-proc", type=int, default=8) - - args = parser.parse_args() - - main(args) diff --git a/fairseq/examples/discriminative_reranking_nmt/tasks/__init__.py b/fairseq/examples/discriminative_reranking_nmt/tasks/__init__.py deleted file mode 100644 index 2d78ca98708121261aa365738a65c051b5b40626..0000000000000000000000000000000000000000 --- a/fairseq/examples/discriminative_reranking_nmt/tasks/__init__.py +++ /dev/null @@ -1,6 +0,0 @@ -from .discriminative_reranking_task import DiscriminativeRerankingNMTTask - - -__all__ = [ - "DiscriminativeRerankingNMTTask", -] diff --git a/fairseq/examples/discriminative_reranking_nmt/tasks/discriminative_reranking_task.py b/fairseq/examples/discriminative_reranking_nmt/tasks/discriminative_reranking_task.py deleted file mode 100644 index 0e7fbba888c1ddd118da8238d644b4ab571177ff..0000000000000000000000000000000000000000 --- a/fairseq/examples/discriminative_reranking_nmt/tasks/discriminative_reranking_task.py +++ /dev/null @@ -1,475 +0,0 @@ -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. - -from dataclasses import dataclass, field - -import itertools -import logging -import os - -import numpy as np -import torch - -from fairseq import metrics -from fairseq.data import ( - ConcatDataset, - ConcatSentencesDataset, - data_utils, - Dictionary, - IdDataset, - indexed_dataset, - NestedDictionaryDataset, - NumSamplesDataset, - NumelDataset, - PrependTokenDataset, - RawLabelDataset, - RightPadDataset, - SortDataset, - TruncateDataset, - TokenBlockDataset, -) -from fairseq.dataclass import ChoiceEnum, FairseqDataclass -from fairseq.tasks import FairseqTask, register_task -from omegaconf import II, MISSING - - -EVAL_BLEU_ORDER = 4 -TARGET_METRIC_CHOICES = ChoiceEnum(["bleu", "ter"]) - -logger = logging.getLogger(__name__) - - -@dataclass -class DiscriminativeRerankingNMTConfig(FairseqDataclass): - data: str = field(default=MISSING, metadata={"help": "path to data directory"}) - num_data_splits: int = field( - default=1, metadata={"help": "total number of data splits"} - ) - no_shuffle: bool = field( - default=False, metadata={"help": "do not shuffle training data"} - ) - max_positions: int = field( - default=512, metadata={"help": "number of positional embeddings to learn"} - ) - include_src: bool = field( - default=False, metadata={"help": "include source sentence"} - ) - mt_beam: int = field(default=50, metadata={"help": "beam size of input hypotheses"}) - eval_target_metric: bool = field( - default=False, - metadata={"help": "evaluation with the target metric during validation"}, - ) - target_metric: TARGET_METRIC_CHOICES = field( - default="bleu", metadata={"help": "name of the target metric to optimize for"} - ) - train_subset: str = field( - default=II("dataset.train_subset"), - metadata={"help": "data subset to use for training (e.g. train, valid, test)"}, - ) - seed: int = field( - default=II("common.seed"), - metadata={"help": "pseudo random number generator seed"}, - ) - - -class RerankerScorer(object): - """Scores the target for a given (source (optional), target) input.""" - - def __init__(self, args, mt_beam): - self.mt_beam = mt_beam - - @torch.no_grad() - def generate(self, models, sample, **kwargs): - """Score a batch of translations.""" - net_input = sample["net_input"] - - assert len(models) == 1, "does not support model ensemble" - model = models[0] - - bs = net_input["src_tokens"].shape[0] - assert ( - model.joint_classification == "none" or bs % self.mt_beam == 0 - ), f"invalid batch size ({bs}) for joint classification with beam size ({self.mt_beam})" - - model.eval() - logits = model(**net_input) - - batch_out = model.sentence_forward(logits, net_input["src_tokens"]) - if model.joint_classification == "sent": - batch_out = model.joint_forward( - batch_out.view(self.mt_beam, bs // self.mt_beam, -1) - ) - scores = model.classification_forward( - batch_out.view(bs, 1, -1) - ) # input: B x T x C - - return scores - - -@register_task( - "discriminative_reranking_nmt", dataclass=DiscriminativeRerankingNMTConfig -) -class DiscriminativeRerankingNMTTask(FairseqTask): - """ - Translation rerank task. - The input can be either (src, tgt) sentence pairs or tgt sentence only. - """ - - cfg: DiscriminativeRerankingNMTConfig - - def __init__(self, cfg: DiscriminativeRerankingNMTConfig, data_dictionary=None): - super().__init__(cfg) - self.dictionary = data_dictionary - self._max_positions = cfg.max_positions - # args.tokens_per_sample = self._max_positions - # self.num_classes = 1 # for model - - @classmethod - def load_dictionary(cls, cfg, filename): - """Load the dictionary from the filename""" - dictionary = Dictionary.load(filename) - dictionary.add_symbol("
-
-
Limitations
- -FSDP currently has several limitations compared to fairseq's default DDP backend (PyTorch DDP): -* while FSDP is full compatible with pointwise Optimizers (e.g., Adam, AdamW, Adadelta, Adamax, SGD, etc.), it is not currently compatible with non-pointwise Optimizers (e.g., Adagrad, Adafactor, LAMB, etc.) -* FSDP depends on flattening the parameters, so models that currently require `--fp16-no-flatten-grads` may not be supported - -See the [fairscale docs](https://fairscale.readthedocs.io/en/latest/api/nn/fsdp_tips.html) for a more detailed -explanation of these and other limitations. - -
How it works
-
-
-
-See the [fairscale docs](https://fairscale.readthedocs.io/en/latest/api/nn/fsdp_tips.html) for a more detailed
-explanation of how FSDP works.
-
-
Example output
- -``` -(...) -2021-03-08 12:29:51 | INFO | fairseq_cli.train | num. model params: 13,110,865,920 (num. trained: 13,110,865,920) -(...) -2021-03-08 12:29:51 | INFO | fairseq_cli.train | training on 1 devices (GPUs/TPUs) -2021-03-08 12:29:51 | INFO | fairseq_cli.train | max tokens per GPU = None and batch size per GPU = 8 -(...) -Adam Optimizer #0 is created with AVX2 arithmetic capability. -Config: alpha=0.000100, betas=(0.900000, 0.980000), weight_decay=0.000000, adam_w=1 -(...) -2021-03-08 12:31:36 | INFO | train_inner | {"epoch": 1, "update": 0.0, "loss": "16.475", "ppl": "91120.8", "wps": "0", "ups": "0", "wpb": "16384", "bsz": "8", "num_updates": "1", "lr": "2e-05", "gnorm": "20.751", "loss_scale": "4", "train_wall": "99", "gb_free": "9.3", "wall": "105"} -2021-03-08 12:32:33 | INFO | train_inner | {"epoch": 1, "update": 0.0, "loss": "16.446", "ppl": "89281.6", "wps": "288.7", "ups": "0.02", "wpb": "16384", "bsz": "8", "num_updates": "2", "lr": "4e-05", "gnorm": "19.777", "loss_scale": "4", "train_wall": "57", "gb_free": "9.3", "wall": "161"} -2021-03-08 12:33:12 | INFO | fairseq.trainer | NOTE: gradient overflow detected, ignoring gradient, setting loss scale to: 2.0 -2021-03-08 12:33:51 | INFO | fairseq.trainer | NOTE: gradient overflow detected, ignoring gradient, setting loss scale to: 1.0 -2021-03-08 12:34:45 | INFO | train_inner | {"epoch": 1, "update": 0.001, "loss": "25.22", "ppl": "3.90691e+07", "wps": "123.4", "ups": "0.01", "wpb": "16384", "bsz": "8", "num_updates": "3", "lr": "6e-05", "gnorm": "131.281", "loss_scale": "1", "train_wall": "133", "gb_free": "9.3", "wall": "294"} -2021-03-08 12:35:43 | INFO | train_inner | {"epoch": 1, "update": 0.001, "loss": "18.079", "ppl": "276809", "wps": "285.5", "ups": "0.02", "wpb": "16384", "bsz": "8", "num_updates": "4", "lr": "8e-05", "gnorm": "13.776", "loss_scale": "1", "train_wall": "57", "gb_free": "9.3", "wall": "351"} -2021-03-08 12:36:35 | INFO | train_inner | {"epoch": 1, "update": 0.001, "loss": "23.729", "ppl": "1.39088e+07", "wps": "316.7", "ups": "0.02", "wpb": "16384", "bsz": "8", "num_updates": "5", "lr": "0.0001", "gnorm": "72.774", "loss_scale": "1", "train_wall": "52", "gb_free": "9.3", "wall": "403"} -2021-03-08 12:37:28 | INFO | train_inner | {"epoch": 1, "update": 0.001, "loss": "20.429", "ppl": "1.41203e+06", "wps": "307.6", "ups": "0.02", "wpb": "16384", "bsz": "8", "num_updates": "6", "lr": "8e-05", "gnorm": "60.846", "loss_scale": "1", "train_wall": "53", "gb_free": "9.3", "wall": "456"} -2021-03-08 12:38:27 | INFO | train_inner | {"epoch": 1, "update": 0.001, "loss": "18.965", "ppl": "511684", "wps": "279.4", "ups": "0.02", "wpb": "16384", "bsz": "8", "num_updates": "7", "lr": "6e-05", "gnorm": "22.687", "loss_scale": "1", "train_wall": "59", "gb_free": "9.3", "wall": "515"} -2021-03-08 12:39:18 | INFO | train_inner | {"epoch": 1, "update": 0.001, "loss": "18.345", "ppl": "332887", "wps": "319.1", "ups": "0.02", "wpb": "16384", "bsz": "8", "num_updates": "8", "lr": "4e-05", "gnorm": "8.451", "loss_scale": "1", "train_wall": "51", "gb_free": "9.3", "wall": "566"} -2021-03-08 12:40:11 | INFO | train_inner | {"epoch": 1, "update": 0.002, "loss": "18.262", "ppl": "314336", "wps": "305.9", "ups": "0.02", "wpb": "16384", "bsz": "8", "num_updates": "9", "lr": "2e-05", "gnorm": "6.457", "loss_scale": "1", "train_wall": "54", "gb_free": "9.3", "wall": "620"} -2021-03-08 12:41:04 | INFO | train_inner | {"epoch": 1, "update": 0.002, "loss": "17.556", "ppl": "192686", "wps": "311.8", "ups": "0.02", "wpb": "16384", "bsz": "8", "num_updates": "10", "lr": "0", "gnorm": "5.796", "loss_scale": "1", "train_wall": "53", "gb_free": "9.3", "wall": "673"} -2021-03-08 12:41:04 | INFO | fairseq_cli.train | Stopping training due to num_updates: 10 >= max_update: 10 -2021-03-08 12:41:04 | INFO | fairseq_cli.train | begin validation on "valid" subset -2021-03-08 12:43:15 | INFO | valid | {"epoch": 1, "valid_loss": "17.953", "valid_ppl": "253807", "valid_wps": "1868.4", "valid_wpb": "15400.2", "valid_bsz": "7.6", "valid_num_updates": "10"} -2021-03-08 12:43:15 | INFO | fairseq_cli.train | end of epoch 1 (average epoch stats below) -2021-03-08 12:43:15 | INFO | train | {"epoch": 1, "train_loss": "19.351", "train_ppl": "668509", "train_wps": "210.9", "train_ups": "0.01", "train_wpb": "16384", "train_bsz": "8", "train_num_updates": "10", "train_lr": "0", "train_gnorm": "36.26", "train_loss_scale": "1", "train_train_wall": "667", "train_gb_free": "9.3", "train_wall": "804"} -2021-03-08 12:43:15 | INFO | fairseq_cli.train | done training in 798.6 seconds -``` - -
Example output
- -``` -(...) -2021-03-08 18:04:09 | INFO | fairseq_cli.train | num. model params: 13,110,865,920 (num. trained: 13,110,865,920) -(...) -2021-03-08 18:04:09 | INFO | fairseq_cli.train | training on 8 devices (GPUs/TPUs) -2021-03-08 18:04:09 | INFO | fairseq_cli.train | max tokens per GPU = None and batch size per GPU = 8 -(...) -Adam Optimizer #0 is created with AVX2 arithmetic capability. -Config: alpha=0.000100, betas=(0.900000, 0.980000), weight_decay=0.000000, adam_w=1 -(...) -2021-03-08 18:05:06 | INFO | train_inner | {"epoch": 1, "update": 0.001, "loss": "16.408", "ppl": "86945.6", "wps": "0", "ups": "0", "wpb": "131072", "bsz": "64", "num_updates": "1", "lr": "2e-05", "gnorm": "18.27", "loss_scale": "4", "train_wall": "47", "gb_free": "9.3", "wall": "56"} -2021-03-08 18:05:45 | INFO | train_inner | {"epoch": 1, "update": 0.002, "loss": "16.352", "ppl": "83644.3", "wps": "3283.4", "ups": "0.03", "wpb": "131072", "bsz": "64", "num_updates": "2", "lr": "4e-05", "gnorm": "18.411", "loss_scale": "4", "train_wall": "40", "gb_free": "9.3", "wall": "96"} -2021-03-08 18:06:21 | INFO | fairseq.trainer | NOTE: gradient overflow detected, ignoring gradient, setting loss scale to: 2.0 -2021-03-08 18:06:56 | INFO | fairseq.trainer | NOTE: gradient overflow detected, ignoring gradient, setting loss scale to: 1.0 -2021-03-08 18:07:37 | INFO | train_inner | {"epoch": 1, "update": 0.006, "loss": "23.682", "ppl": "1.34537e+07", "wps": "1176.6", "ups": "0.01", "wpb": "131072", "bsz": "64", "num_updates": "3", "lr": "6e-05", "gnorm": "119.682", "loss_scale": "1", "train_wall": "111", "gb_free": "9.3", "wall": "208"} -2021-03-08 18:08:18 | INFO | train_inner | {"epoch": 1, "update": 0.007, "loss": "18.988", "ppl": "519921", "wps": "3189.1", "ups": "0.02", "wpb": "131072", "bsz": "64", "num_updates": "4", "lr": "8e-05", "gnorm": "14.934", "loss_scale": "1", "train_wall": "41", "gb_free": "9.3", "wall": "249"} -2021-03-08 18:08:59 | INFO | train_inner | {"epoch": 1, "update": 0.008, "loss": "20.08", "ppl": "1.10798e+06", "wps": "3223.1", "ups": "0.02", "wpb": "131072", "bsz": "64", "num_updates": "5", "lr": "0.0001", "gnorm": "59.92", "loss_scale": "1", "train_wall": "41", "gb_free": "9.3", "wall": "289"} -2021-03-08 18:09:39 | INFO | train_inner | {"epoch": 1, "update": 0.009, "loss": "18.323", "ppl": "327980", "wps": "3256.6", "ups": "0.02", "wpb": "131072", "bsz": "64", "num_updates": "6", "lr": "8e-05", "gnorm": "37.425", "loss_scale": "1", "train_wall": "40", "gb_free": "9.3", "wall": "330"} -2021-03-08 18:10:20 | INFO | train_inner | {"epoch": 1, "update": 0.01, "loss": "17.264", "ppl": "157354", "wps": "3188.7", "ups": "0.02", "wpb": "131072", "bsz": "64", "num_updates": "7", "lr": "6e-05", "gnorm": "10.824", "loss_scale": "1", "train_wall": "41", "gb_free": "9.3", "wall": "371"} -2021-03-08 18:11:01 | INFO | train_inner | {"epoch": 1, "update": 0.011, "loss": "16.794", "ppl": "113647", "wps": "3230", "ups": "0.02", "wpb": "131072", "bsz": "64", "num_updates": "8", "lr": "4e-05", "gnorm": "5.616", "loss_scale": "1", "train_wall": "41", "gb_free": "9.3", "wall": "411"} -2021-03-08 18:11:39 | INFO | train_inner | {"epoch": 1, "update": 0.012, "loss": "16.706", "ppl": "106938", "wps": "3384", "ups": "0.03", "wpb": "131072", "bsz": "64", "num_updates": "9", "lr": "2e-05", "gnorm": "5.318", "loss_scale": "1", "train_wall": "39", "gb_free": "9.3", "wall": "450"} -2021-03-08 18:12:19 | INFO | train_inner | {"epoch": 1, "update": 0.013, "loss": "16.548", "ppl": "95796.2", "wps": "3274.4", "ups": "0.02", "wpb": "131072", "bsz": "64", "num_updates": "10", "lr": "0", "gnorm": "5.22", "loss_scale": "1", "train_wall": "40", "gb_free": "9.3", "wall": "490"} -2021-03-08 18:12:19 | INFO | fairseq_cli.train | Stopping training due to num_updates: 10 >= max_update: 10 -2021-03-08 18:12:19 | INFO | fairseq_cli.train | begin validation on "valid" subset -2021-03-08 18:12:45 | INFO | valid | {"epoch": 1, "valid_loss": "16.624", "valid_ppl": "101000", "valid_wps": "10855.9", "valid_wpb": "123202", "valid_bsz": "60.5", "valid_num_updates": "10"} -2021-03-08 18:12:45 | INFO | fairseq_cli.train | end of epoch 1 (average epoch stats below) -2021-03-08 18:12:45 | INFO | train | {"epoch": 1, "train_loss": "18.114", "train_ppl": "283776", "train_wps": "2567.8", "train_ups": "0.02", "train_wpb": "131072", "train_bsz": "64", "train_num_updates": "10", "train_lr": "0", "train_gnorm": "29.562", "train_loss_scale": "1", "train_train_wall": "480", "train_gb_free": "9.3", "train_wall": "516"} -2021-03-08 18:12:45 | INFO | fairseq_cli.train | done training in 509.9 seconds -``` - -
([Baevski and Auli, 2018](https://arxiv.org/abs/1809.10853)) | 1026M | [Google Billion Words](https://github.com/ciprian-chelba/1-billion-word-language-modeling-benchmark) | [download (.tar.bz2)](https://dl.fbaipublicfiles.com/fairseq/models/lm/adaptive_lm_gbw_huge.tar.bz2) -Adaptive Inputs
([Baevski and Auli, 2018](https://arxiv.org/abs/1809.10853)) | 247M | [WikiText-103](https://blog.einstein.ai/the-wikitext-long-term-dependency-language-modeling-dataset/) | [download (.tar.bz2)](https://dl.fbaipublicfiles.com/fairseq/models/lm/adaptive_lm_wiki103.v2.tar.bz2) - -## Training an LM with adaptive inputs - -First, see the general [language modeling README](README.md) for instructions on -preprocessing the WikiText-103 data. - -Then use the following training command to train a model with adaptive inputs -using the `transformer_lm_wiki103` model architecture: -```bash -fairseq-train --task language_modeling \ - data-bin/wikitext-103 \ - --save-dir checkpoints/transformer_wikitext-103 \ - --arch transformer_lm_wiki103 \ - --max-update 286000 --lr 1.0 --t-mult 2 --lr-period-updates 270000 --lr-scheduler cosine --lr-shrink 0.75 \ - --warmup-updates 16000 --warmup-init-lr 1e-07 --stop-min-lr 1e-09 --optimizer nag --min-lr 0.0001 --clip-norm 0.1 \ - --criterion adaptive_loss --max-tokens 3072 --update-freq 3 --tokens-per-sample 3072 --seed 1 \ - --sample-break-mode none --skip-invalid-size-inputs-valid-test --ddp-backend=legacy_ddp -``` - -## Citation - -```bibtex -@inproceedings{ - baevski2018adaptive, - title={Adaptive Input Representations for Neural Language Modeling}, - author={Alexei Baevski and Michael Auli}, - booktitle={International Conference on Learning Representations}, - year={2019}, - url={https://openreview.net/forum?id=ByxZX20qFQ}, -} -``` diff --git a/fairseq/examples/language_model/README.conv.md b/fairseq/examples/language_model/README.conv.md deleted file mode 100644 index 1ff8635906cf278208be4714e0ef805a6a6b4da1..0000000000000000000000000000000000000000 --- a/fairseq/examples/language_model/README.conv.md +++ /dev/null @@ -1,40 +0,0 @@ -# Language Modeling with Gated Convolutional Networks (Dauphin et al., 2017) - -## Example usage - -First download and preprocess the data following the main [language modeling README](README.md). - -Then to train a convolutional LM using the `fconv_lm_dauphin_wikitext103` -architecture: -```bash -fairseq-train --task language_modeling \ - data-bin/wikitext-103 \ - --save-dir checkpoints/fconv_wikitext-103 \ - --arch fconv_lm_dauphin_wikitext103 \ - --adaptive-softmax-cutoff 10000,20000,200000 \ - --dropout 0.2 \ - --criterion adaptive_loss \ - --optimizer nag --clip-norm 0.1 --weight-decay 5e-06 \ - --lr 1.0 --lr-scheduler reduce_lr_on_plateau --lr-shrink 0.5 \ - --max-tokens 1024 --tokens-per-sample 1024 \ - --ddp-backend legacy_ddp \ - --max-epoch 35 -``` - -And evaluate with: -```bash -fairseq-eval-lm data-bin/wikitext-103 --path checkpoints/fconv_wiki103/checkpoint_best.pt -``` - -## Citation - -```bibtex -@inproceedings{dauphin2017language, - title={Language Modeling with Gated Convolutional Networks}, - author={Dauphin, Yann N and Fan, Angela and Auli, Michael and Grangier, David}, - booktitle={Proceedings of the 34th International Conference on Machine Learning-Volume 70}, - pages={933--941}, - year={2017}, - organization={JMLR} -} -``` diff --git a/fairseq/examples/language_model/README.md b/fairseq/examples/language_model/README.md deleted file mode 100644 index e78ea48e08dc99b69751923762107a8f8a9a5e3e..0000000000000000000000000000000000000000 --- a/fairseq/examples/language_model/README.md +++ /dev/null @@ -1,123 +0,0 @@ -# Neural Language Modeling - -## Pre-trained models - -Model | Description | Dataset | Download ----|---|---|--- -`transformer_lm.gbw.adaptive_huge` | Adaptive Inputs
([Baevski and Auli, 2018](https://arxiv.org/abs/1809.10853))
1026M params | [Google Billion Words](https://github.com/ciprian-chelba/1-billion-word-language-modeling-benchmark) | [download (.tar.bz2)](https://dl.fbaipublicfiles.com/fairseq/models/lm/adaptive_lm_gbw_huge.tar.bz2) -`transformer_lm.wiki103.adaptive` | Adaptive Inputs
([Baevski and Auli, 2018](https://arxiv.org/abs/1809.10853))
247M params | [WikiText-103](https://blog.einstein.ai/the-wikitext-long-term-dependency-language-modeling-dataset) | [download (.tar.bz2)](https://dl.fbaipublicfiles.com/fairseq/models/lm/adaptive_lm_wiki103.v2.tar.bz2) -`transformer_lm.wmt19.en` | English LM
([Ng et al., 2019](https://arxiv.org/abs/1907.06616)) | [WMT News Crawl](http://data.statmt.org/news-crawl/) | [download (.tar.gz)](https://dl.fbaipublicfiles.com/fairseq/models/lm/wmt19.en.tar.gz) -`transformer_lm.wmt19.de` | German LM
([Ng et al., 2019](https://arxiv.org/abs/1907.06616)) | [WMT News Crawl](http://data.statmt.org/news-crawl/) | [download (.tar.gz)](https://dl.fbaipublicfiles.com/fairseq/models/lm/wmt19.de.tar.gz) -`transformer_lm.wmt19.ru` | Russian LM
([Ng et al., 2019](https://arxiv.org/abs/1907.06616)) | [WMT News Crawl](http://data.statmt.org/news-crawl/) | [download (.tar.gz)](https://dl.fbaipublicfiles.com/fairseq/models/lm/wmt19.ru.tar.gz) - -## Example usage - -We require a few additional Python dependencies for preprocessing: -```bash -pip install fastBPE sacremoses -``` - -To sample from a language model using PyTorch Hub: -```python -import torch - -# List available models -torch.hub.list('pytorch/fairseq') # [..., 'transformer_lm.wmt19.en', ...] - -# Load an English LM trained on WMT'19 News Crawl data -en_lm = torch.hub.load('pytorch/fairseq', 'transformer_lm.wmt19.en', tokenizer='moses', bpe='fastbpe') -en_lm.eval() # disable dropout - -# Move model to GPU -en_lm.cuda() - -# Sample from the language model -en_lm.sample('Barack Obama', beam=1, sampling=True, sampling_topk=10, temperature=0.8) -# "Barack Obama is coming to Sydney and New Zealand (...)" - -# Compute perplexity for a sequence -en_lm.score('Barack Obama is coming to Sydney and New Zealand')['positional_scores'].mean().neg().exp() -# tensor(15.1474) - -# The same interface can be used with custom models as well -from fairseq.models.transformer_lm import TransformerLanguageModel -custom_lm = TransformerLanguageModel.from_pretrained('/path/to/model/dir', 'checkpoint100.pt', tokenizer='moses', bpe='fastbpe') -custom_lm.sample('Barack Obama', beam=5) -# "Barack Obama (...)" -``` - -## Training a transformer language model with the CLI tools - -### 1) Preprocess the data - -First download and prepare the [WikiText-103 dataset](https://www.salesforce.com/products/einstein/ai-research/the-wikitext-dependency-language-modeling-dataset/): -```bash -cd examples/language_model/ -bash prepare-wikitext-103.sh -cd ../.. -``` - -Next preprocess/binarize the data: -```bash -TEXT=examples/language_model/wikitext-103 -fairseq-preprocess \ - --only-source \ - --trainpref $TEXT/wiki.train.tokens \ - --validpref $TEXT/wiki.valid.tokens \ - --testpref $TEXT/wiki.test.tokens \ - --destdir data-bin/wikitext-103 \ - --workers 20 -``` - -### 2) Train a language model - -Next we'll train a basic transformer language model on wikitext-103. For more -advanced usage, see the [adaptive inputs README](README.adaptive_inputs.md). - -To train a basic LM (assumes 2 GPUs): -``` -$ fairseq-train --task language_modeling \ - data-bin/wikitext-103 \ - --save-dir checkpoints/transformer_wikitext-103 \ - --arch transformer_lm --share-decoder-input-output-embed \ - --dropout 0.1 \ - --optimizer adam --adam-betas '(0.9, 0.98)' --weight-decay 0.01 --clip-norm 0.0 \ - --lr 0.0005 --lr-scheduler inverse_sqrt --warmup-updates 4000 --warmup-init-lr 1e-07 \ - --tokens-per-sample 512 --sample-break-mode none \ - --max-tokens 2048 --update-freq 16 \ - --fp16 \ - --max-update 50000 -``` - -If you run out of memory, try reducing `--max-tokens` (max number of tokens per -batch) or `--tokens-per-sample` (max sequence length). You can also adjust -`--update-freq` to accumulate gradients and simulate training on a different -number of GPUs. - -### 3) Evaluate - -```bash -fairseq-eval-lm data-bin/wikitext-103 \ - --path checkpoints/transformer_wiki103/checkpoint_best.pt \ - --batch-size 2 \ - --tokens-per-sample 512 \ - --context-window 400 -# | Evaluated 245569 tokens in 56.1s (4379.02 tokens/s) -# | Loss: 3.4164, Perplexity: 30.46 -``` - -*Note:* The `--context-window` option controls how much context is provided to -each token when computing perplexity. When the window size is 0, the dataset is -chunked into segments of length 512 and perplexity is computed over each segment -normally. However, this results in worse (higher) perplexity since tokens that -appear earlier in each segment have less conditioning. When the maximum window -size is used (511 in this case), then we compute perplexity for each token -fully conditioned on 511 tokens of context. This slows down evaluation -significantly, since we must run a separate forward pass for every token in the -dataset, but results in better (lower) perplexity. - - -## Convolutional language models - -Please see the [convolutional LM README](README.conv.md) for instructions on -training convolutional language models. diff --git a/fairseq/examples/language_model/prepare-wikitext-103.sh b/fairseq/examples/language_model/prepare-wikitext-103.sh deleted file mode 100644 index 751302156f0a6829af9c2ee5e0e2ca62c2cd4187..0000000000000000000000000000000000000000 --- a/fairseq/examples/language_model/prepare-wikitext-103.sh +++ /dev/null @@ -1,33 +0,0 @@ -#!/bin/bash -# Adapted from https://github.com/facebookresearch/MIXER/blob/master/prepareData.sh - -URLS=( - "https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-103-v1.zip" -) -FILES=( - "wikitext-103-v1.zip" -) - -for ((i=0;i<${#URLS[@]};++i)); do - file=${FILES[i]} - if [ -f $file ]; then - echo "$file already exists, skipping download" - else - url=${URLS[i]} - wget "$url" - if [ -f $file ]; then - echo "$url successfully downloaded." - else - echo "$url not successfully downloaded." - exit -1 - fi - if [ ${file: -4} == ".tgz" ]; then - tar zxvf $file - elif [ ${file: -4} == ".tar" ]; then - tar xvf $file - elif [ ${file: -4} == ".zip" ]; then - unzip $file - fi - fi -done -cd .. diff --git a/fairseq/examples/laser/README.md b/fairseq/examples/laser/README.md deleted file mode 100644 index 66acada04f58fa235cd312753f144f6f1e5f4a33..0000000000000000000000000000000000000000 --- a/fairseq/examples/laser/README.md +++ /dev/null @@ -1,144 +0,0 @@ -# LASER Language-Agnostic SEntence Representations - -LASER is a library to calculate and use multilingual sentence embeddings. - -You can find more information about LASER and how to use it on the official [LASER repository](https://github.com/facebookresearch/LASER). - -This folder contains source code for training LASER embeddings. - - -## Prepare data and configuration file - -Binarize your data with fairseq, as described [here](https://fairseq.readthedocs.io/en/latest/getting_started.html#data-pre-processing). - -Create a json config file with this format: -``` -{ - "src_vocab": "/path/to/spm.src.cvocab", - "tgt_vocab": "/path/to/spm.tgt.cvocab", - "train": [ - { - "type": "translation", - "id": 0, - "src": "/path/to/srclang1-tgtlang0/train.srclang1", - "tgt": "/path/to/srclang1-tgtlang0/train.tgtlang0" - }, - { - "type": "translation", - "id": 1, - "src": "/path/to/srclang1-tgtlang1/train.srclang1", - "tgt": "/path/to/srclang1-tgtlang1/train.tgtlang1" - }, - { - "type": "translation", - "id": 0, - "src": "/path/to/srclang2-tgtlang0/train.srclang2", - "tgt": "/path/to/srclang2-tgtlang0/train.tgtlang0" - }, - { - "type": "translation", - "id": 1, - "src": "/path/to/srclang2-tgtlang1/train.srclang2", - "tgt": "/path/to/srclang2-tgtlang1/train.tgtlang1" - }, - ... - ], - "valid": [ - { - "type": "translation", - "id": 0, - "src": "/unused", - "tgt": "/unused" - } - ] -} -``` -where paths are paths to binarized indexed fairseq dataset files. -`id` represents the target language id. - - -## Training Command Line Example - -``` -fairseq-train \ - /path/to/configfile_described_above.json \ - --user-dir examples/laser/laser_src \ - --log-interval 100 --log-format simple \ - --task laser --arch laser_lstm \ - --save-dir . \ - --optimizer adam \ - --lr 0.001 \ - --lr-scheduler inverse_sqrt \ - --clip-norm 5 \ - --warmup-updates 90000 \ - --update-freq 2 \ - --dropout 0.0 \ - --encoder-dropout-out 0.1 \ - --max-tokens 2000 \ - --max-epoch 50 \ - --encoder-bidirectional \ - --encoder-layers 5 \ - --encoder-hidden-size 512 \ - --decoder-layers 1 \ - --decoder-hidden-size 2048 \ - --encoder-embed-dim 320 \ - --decoder-embed-dim 320 \ - --decoder-lang-embed-dim 32 \ - --warmup-init-lr 0.001 \ - --disable-validation -``` - - -## Applications - -We showcase several applications of multilingual sentence embeddings -with code to reproduce our results (in the directory "tasks"). - -* [**Cross-lingual document classification**](https://github.com/facebookresearch/LASER/tree/master/tasks/mldoc) using the - [*MLDoc*](https://github.com/facebookresearch/MLDoc) corpus [2,6] -* [**WikiMatrix**](https://github.com/facebookresearch/LASER/tree/master/tasks/WikiMatrix) - Mining 135M Parallel Sentences in 1620 Language Pairs from Wikipedia [7] -* [**Bitext mining**](https://github.com/facebookresearch/LASER/tree/master/tasks/bucc) using the - [*BUCC*](https://comparable.limsi.fr/bucc2018/bucc2018-task.html) corpus [3,5] -* [**Cross-lingual NLI**](https://github.com/facebookresearch/LASER/tree/master/tasks/xnli) - using the [*XNLI*](https://www.nyu.edu/projects/bowman/xnli/) corpus [4,5,6] -* [**Multilingual similarity search**](https://github.com/facebookresearch/LASER/tree/master/tasks/similarity) [1,6] -* [**Sentence embedding of text files**](https://github.com/facebookresearch/LASER/tree/master/tasks/embed) - example how to calculate sentence embeddings for arbitrary text files in any of the supported language. - -**For all tasks, we use exactly the same multilingual encoder, without any task specific optimization or fine-tuning.** - - - -## References - -[1] Holger Schwenk and Matthijs Douze, - [*Learning Joint Multilingual Sentence Representations with Neural Machine Translation*](https://aclanthology.info/papers/W17-2619/w17-2619), - ACL workshop on Representation Learning for NLP, 2017 - -[2] Holger Schwenk and Xian Li, - [*A Corpus for Multilingual Document Classification in Eight Languages*](http://www.lrec-conf.org/proceedings/lrec2018/pdf/658.pdf), - LREC, pages 3548-3551, 2018. - -[3] Holger Schwenk, - [*Filtering and Mining Parallel Data in a Joint Multilingual Space*](http://aclweb.org/anthology/P18-2037) - ACL, July 2018 - -[4] Alexis Conneau, Guillaume Lample, Ruty Rinott, Adina Williams, Samuel R. Bowman, Holger Schwenk and Veselin Stoyanov, - [*XNLI: Cross-lingual Sentence Understanding through Inference*](https://aclweb.org/anthology/D18-1269), - EMNLP, 2018. - -[5] Mikel Artetxe and Holger Schwenk, - [*Margin-based Parallel Corpus Mining with Multilingual Sentence Embeddings*](https://arxiv.org/abs/1811.01136) - arXiv, Nov 3 2018. - -[6] Mikel Artetxe and Holger Schwenk, - [*Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond*](https://arxiv.org/abs/1812.10464) - arXiv, Dec 26 2018. - -[7] Holger Schwenk, Vishrav Chaudhary, Shuo Sun, Hongyu Gong and Paco Guzman, - [*WikiMatrix: Mining 135M Parallel Sentences in 1620 Language Pairs from Wikipedia*](https://arxiv.org/abs/1907.05791) - arXiv, July 11 2019. - -[8] Holger Schwenk, Guillaume Wenzek, Sergey Edunov, Edouard Grave and Armand Joulin - [*CCMatrix: Mining Billions of High-Quality Parallel Sentences on the WEB*](https://arxiv.org/abs/1911.04944) diff --git a/fairseq/examples/laser/laser_src/__init__.py b/fairseq/examples/laser/laser_src/__init__.py deleted file mode 100644 index 9ffbd656d8786e421008fb4cb0d1d8911dc8330c..0000000000000000000000000000000000000000 --- a/fairseq/examples/laser/laser_src/__init__.py +++ /dev/null @@ -1,8 +0,0 @@ -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. - -from .laser_task import * # noqa -from .laser_lstm import * # noqa -from .laser_transformer import * # noqa diff --git a/fairseq/examples/laser/laser_src/laser_lstm.py b/fairseq/examples/laser/laser_src/laser_lstm.py deleted file mode 100644 index 10df90e002d5a7dd74a571dbc3b328c130c57a0a..0000000000000000000000000000000000000000 --- a/fairseq/examples/laser/laser_src/laser_lstm.py +++ /dev/null @@ -1,585 +0,0 @@ -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. - -import torch -import torch.nn as nn -import torch.nn.functional as F - -from fairseq import options, utils - -from fairseq.models import ( - FairseqEncoder, - FairseqIncrementalDecoder, - FairseqEncoderDecoderModel, - register_model, - register_model_architecture, -) - - -@register_model("laser_lstm") -class LSTMModel(FairseqEncoderDecoderModel): - def __init__(self, encoder, decoder): - super().__init__(encoder, decoder) - - def forward( - self, - src_tokens, - src_lengths, - prev_output_tokens=None, - tgt_tokens=None, - tgt_lengths=None, - target_language_id=None, - dataset_name="", - ): - assert target_language_id is not None - - src_encoder_out = self.encoder(src_tokens, src_lengths, dataset_name) - return self.decoder( - prev_output_tokens, src_encoder_out, lang_id=target_language_id - ) - - @staticmethod - def add_args(parser): - """Add model-specific arguments to the parser.""" - parser.add_argument( - "--dropout", - default=0.1, - type=float, - metavar="D", - help="dropout probability", - ) - parser.add_argument( - "--encoder-embed-dim", - type=int, - metavar="N", - help="encoder embedding dimension", - ) - parser.add_argument( - "--encoder-embed-path", - default=None, - type=str, - metavar="STR", - help="path to pre-trained encoder embedding", - ) - parser.add_argument( - "--encoder-hidden-size", type=int, metavar="N", help="encoder hidden size" - ) - parser.add_argument( - "--encoder-layers", type=int, metavar="N", help="number of encoder layers" - ) - parser.add_argument( - "--encoder-bidirectional", - action="store_true", - help="make all layers of encoder bidirectional", - ) - parser.add_argument( - "--decoder-embed-dim", - type=int, - metavar="N", - help="decoder embedding dimension", - ) - parser.add_argument( - "--decoder-embed-path", - default=None, - type=str, - metavar="STR", - help="path to pre-trained decoder embedding", - ) - parser.add_argument( - "--decoder-hidden-size", type=int, metavar="N", help="decoder hidden size" - ) - parser.add_argument( - "--decoder-layers", type=int, metavar="N", help="number of decoder layers" - ) - parser.add_argument( - "--decoder-out-embed-dim", - type=int, - metavar="N", - help="decoder output embedding dimension", - ) - parser.add_argument( - "--decoder-zero-init", - type=str, - metavar="BOOL", - help="initialize the decoder hidden/cell state to zero", - ) - parser.add_argument( - "--decoder-lang-embed-dim", - type=int, - metavar="N", - help="decoder language embedding dimension", - ) - parser.add_argument( - "--fixed-embeddings", - action="store_true", - help="keep embeddings fixed (ENCODER ONLY)", - ) # TODO Also apply to decoder embeddings? - - # Granular dropout settings (if not specified these default to --dropout) - parser.add_argument( - "--encoder-dropout-in", - type=float, - metavar="D", - help="dropout probability for encoder input embedding", - ) - parser.add_argument( - "--encoder-dropout-out", - type=float, - metavar="D", - help="dropout probability for encoder output", - ) - parser.add_argument( - "--decoder-dropout-in", - type=float, - metavar="D", - help="dropout probability for decoder input embedding", - ) - parser.add_argument( - "--decoder-dropout-out", - type=float, - metavar="D", - help="dropout probability for decoder output", - ) - - @classmethod - def build_model(cls, args, task): - """Build a new model instance.""" - # make sure that all args are properly defaulted (in case there are any new ones) - base_architecture(args) - - def load_pretrained_embedding_from_file(embed_path, dictionary, embed_dim): - num_embeddings = len(dictionary) - padding_idx = dictionary.pad() - embed_tokens = Embedding(num_embeddings, embed_dim, padding_idx) - embed_dict = utils.parse_embedding(embed_path) - utils.print_embed_overlap(embed_dict, dictionary) - return utils.load_embedding(embed_dict, dictionary, embed_tokens) - - pretrained_encoder_embed = None - if args.encoder_embed_path: - pretrained_encoder_embed = load_pretrained_embedding_from_file( - args.encoder_embed_path, task.source_dictionary, args.encoder_embed_dim - ) - pretrained_decoder_embed = None - if args.decoder_embed_path: - pretrained_decoder_embed = load_pretrained_embedding_from_file( - args.decoder_embed_path, task.target_dictionary, args.decoder_embed_dim - ) - - num_langs = task.num_tasks if hasattr(task, "num_tasks") else 0 - - encoder = LSTMEncoder( - dictionary=task.source_dictionary, - embed_dim=args.encoder_embed_dim, - hidden_size=args.encoder_hidden_size, - num_layers=args.encoder_layers, - dropout_in=args.encoder_dropout_in, - dropout_out=args.encoder_dropout_out, - bidirectional=args.encoder_bidirectional, - pretrained_embed=pretrained_encoder_embed, - fixed_embeddings=args.fixed_embeddings, - ) - decoder = LSTMDecoder( - dictionary=task.target_dictionary, - embed_dim=args.decoder_embed_dim, - hidden_size=args.decoder_hidden_size, - out_embed_dim=args.decoder_out_embed_dim, - num_layers=args.decoder_layers, - dropout_in=args.decoder_dropout_in, - dropout_out=args.decoder_dropout_out, - zero_init=options.eval_bool(args.decoder_zero_init), - encoder_embed_dim=args.encoder_embed_dim, - encoder_output_units=encoder.output_units, - pretrained_embed=pretrained_decoder_embed, - num_langs=num_langs, - lang_embed_dim=args.decoder_lang_embed_dim, - ) - return cls(encoder, decoder) - - -class LSTMEncoder(FairseqEncoder): - """LSTM encoder.""" - - def __init__( - self, - dictionary, - embed_dim=512, - hidden_size=512, - num_layers=1, - dropout_in=0.1, - dropout_out=0.1, - bidirectional=False, - left_pad=True, - pretrained_embed=None, - padding_value=0.0, - fixed_embeddings=False, - ): - super().__init__(dictionary) - self.num_layers = num_layers - self.dropout_in = dropout_in - self.dropout_out = dropout_out - self.bidirectional = bidirectional - self.hidden_size = hidden_size - - num_embeddings = len(dictionary) - self.padding_idx = dictionary.pad() - if pretrained_embed is None: - self.embed_tokens = Embedding(num_embeddings, embed_dim, self.padding_idx) - else: - self.embed_tokens = pretrained_embed - if fixed_embeddings: - self.embed_tokens.weight.requires_grad = False - - self.lstm = LSTM( - input_size=embed_dim, - hidden_size=hidden_size, - num_layers=num_layers, - dropout=self.dropout_out if num_layers > 1 else 0.0, - bidirectional=bidirectional, - ) - self.left_pad = left_pad - self.padding_value = padding_value - - self.output_units = hidden_size - if bidirectional: - self.output_units *= 2 - - def forward(self, src_tokens, src_lengths, dataset_name): - if self.left_pad: - # convert left-padding to right-padding - src_tokens = utils.convert_padding_direction( - src_tokens, - self.padding_idx, - left_to_right=True, - ) - - bsz, seqlen = src_tokens.size() - - # embed tokens - x = self.embed_tokens(src_tokens) - x = F.dropout(x, p=self.dropout_in, training=self.training) - - # B x T x C -> T x B x C - x = x.transpose(0, 1) - - # pack embedded source tokens into a PackedSequence - try: - packed_x = nn.utils.rnn.pack_padded_sequence(x, src_lengths.data.tolist()) - except BaseException: - raise Exception(f"Packing failed in dataset {dataset_name}") - - # apply LSTM - if self.bidirectional: - state_size = 2 * self.num_layers, bsz, self.hidden_size - else: - state_size = self.num_layers, bsz, self.hidden_size - h0 = x.data.new(*state_size).zero_() - c0 = x.data.new(*state_size).zero_() - packed_outs, (final_hiddens, final_cells) = self.lstm(packed_x, (h0, c0)) - - # unpack outputs and apply dropout - x, _ = nn.utils.rnn.pad_packed_sequence( - packed_outs, padding_value=self.padding_value - ) - x = F.dropout(x, p=self.dropout_out, training=self.training) - assert list(x.size()) == [seqlen, bsz, self.output_units] - - if self.bidirectional: - - def combine_bidir(outs): - return torch.cat( - [ - torch.cat([outs[2 * i], outs[2 * i + 1]], dim=0).view( - 1, bsz, self.output_units - ) - for i in range(self.num_layers) - ], - dim=0, - ) - - final_hiddens = combine_bidir(final_hiddens) - final_cells = combine_bidir(final_cells) - - encoder_padding_mask = src_tokens.eq(self.padding_idx).t() - - # Set padded outputs to -inf so they are not selected by max-pooling - padding_mask = src_tokens.eq(self.padding_idx).t().unsqueeze(-1) - if padding_mask.any(): - x = x.float().masked_fill_(padding_mask, float("-inf")).type_as(x) - - # Build the sentence embedding by max-pooling over the encoder outputs - sentemb = x.max(dim=0)[0] - - return { - "sentemb": sentemb, - "encoder_out": (x, final_hiddens, final_cells), - "encoder_padding_mask": encoder_padding_mask - if encoder_padding_mask.any() - else None, - } - - def reorder_encoder_out(self, encoder_out_dict, new_order): - encoder_out_dict["sentemb"] = encoder_out_dict["sentemb"].index_select( - 0, new_order - ) - encoder_out_dict["encoder_out"] = tuple( - eo.index_select(1, new_order) for eo in encoder_out_dict["encoder_out"] - ) - if encoder_out_dict["encoder_padding_mask"] is not None: - encoder_out_dict["encoder_padding_mask"] = encoder_out_dict[ - "encoder_padding_mask" - ].index_select(1, new_order) - return encoder_out_dict - - def max_positions(self): - """Maximum input length supported by the encoder.""" - return int(1e5) # an arbitrary large number - - -class LSTMDecoder(FairseqIncrementalDecoder): - """LSTM decoder.""" - - def __init__( - self, - dictionary, - embed_dim=512, - hidden_size=512, - out_embed_dim=512, - num_layers=1, - dropout_in=0.1, - dropout_out=0.1, - zero_init=False, - encoder_embed_dim=512, - encoder_output_units=512, - pretrained_embed=None, - num_langs=1, - lang_embed_dim=0, - ): - super().__init__(dictionary) - self.dropout_in = dropout_in - self.dropout_out = dropout_out - self.hidden_size = hidden_size - - num_embeddings = len(dictionary) - padding_idx = dictionary.pad() - if pretrained_embed is None: - self.embed_tokens = Embedding(num_embeddings, embed_dim, padding_idx) - else: - self.embed_tokens = pretrained_embed - - self.layers = nn.ModuleList( - [ - LSTMCell( - input_size=encoder_output_units + embed_dim + lang_embed_dim - if layer == 0 - else hidden_size, - hidden_size=hidden_size, - ) - for layer in range(num_layers) - ] - ) - if hidden_size != out_embed_dim: - self.additional_fc = Linear(hidden_size, out_embed_dim) - self.fc_out = Linear(out_embed_dim, num_embeddings, dropout=dropout_out) - - if zero_init: - self.sentemb2init = None - else: - self.sentemb2init = Linear( - encoder_output_units, 2 * num_layers * hidden_size - ) - - if lang_embed_dim == 0: - self.embed_lang = None - else: - self.embed_lang = nn.Embedding(num_langs, lang_embed_dim) - nn.init.uniform_(self.embed_lang.weight, -0.1, 0.1) - - def forward( - self, prev_output_tokens, encoder_out_dict, incremental_state=None, lang_id=0 - ): - sentemb = encoder_out_dict["sentemb"] - encoder_out = encoder_out_dict["encoder_out"] - - if incremental_state is not None: - prev_output_tokens = prev_output_tokens[:, -1:] - bsz, seqlen = prev_output_tokens.size() - - # get outputs from encoder - encoder_outs, _, _ = encoder_out[:3] - srclen = encoder_outs.size(0) - - # embed tokens - x = self.embed_tokens(prev_output_tokens) - x = F.dropout(x, p=self.dropout_in, training=self.training) - - # embed language identifier - if self.embed_lang is not None: - lang_ids = prev_output_tokens.data.new_full((bsz,), lang_id) - langemb = self.embed_lang(lang_ids) - # TODO Should we dropout here??? - - # B x T x C -> T x B x C - x = x.transpose(0, 1) - - # initialize previous states (or get from cache during incremental generation) - cached_state = utils.get_incremental_state( - self, incremental_state, "cached_state" - ) - if cached_state is not None: - prev_hiddens, prev_cells, input_feed = cached_state - else: - num_layers = len(self.layers) - if self.sentemb2init is None: - prev_hiddens = [ - x.data.new(bsz, self.hidden_size).zero_() for i in range(num_layers) - ] - prev_cells = [ - x.data.new(bsz, self.hidden_size).zero_() for i in range(num_layers) - ] - else: - init = self.sentemb2init(sentemb) - prev_hiddens = [ - init[:, (2 * i) * self.hidden_size : (2 * i + 1) * self.hidden_size] - for i in range(num_layers) - ] - prev_cells = [ - init[ - :, - (2 * i + 1) * self.hidden_size : (2 * i + 2) * self.hidden_size, - ] - for i in range(num_layers) - ] - input_feed = x.data.new(bsz, self.hidden_size).zero_() - - attn_scores = x.data.new(srclen, seqlen, bsz).zero_() - outs = [] - for j in range(seqlen): - if self.embed_lang is None: - input = torch.cat((x[j, :, :], sentemb), dim=1) - else: - input = torch.cat((x[j, :, :], sentemb, langemb), dim=1) - - for i, rnn in enumerate(self.layers): - # recurrent cell - hidden, cell = rnn(input, (prev_hiddens[i], prev_cells[i])) - - # hidden state becomes the input to the next layer - input = F.dropout(hidden, p=self.dropout_out, training=self.training) - - # save state for next time step - prev_hiddens[i] = hidden - prev_cells[i] = cell - - out = hidden - out = F.dropout(out, p=self.dropout_out, training=self.training) - - # input feeding - input_feed = out - - # save final output - outs.append(out) - - # cache previous states (no-op except during incremental generation) - utils.set_incremental_state( - self, - incremental_state, - "cached_state", - (prev_hiddens, prev_cells, input_feed), - ) - - # collect outputs across time steps - x = torch.cat(outs, dim=0).view(seqlen, bsz, self.hidden_size) - - # T x B x C -> B x T x C - x = x.transpose(1, 0) - - # srclen x tgtlen x bsz -> bsz x tgtlen x srclen - attn_scores = attn_scores.transpose(0, 2) - - # project back to size of vocabulary - if hasattr(self, "additional_fc"): - x = self.additional_fc(x) - x = F.dropout(x, p=self.dropout_out, training=self.training) - x = self.fc_out(x) - - return x, attn_scores - - def reorder_incremental_state(self, incremental_state, new_order): - super().reorder_incremental_state(incremental_state, new_order) - cached_state = utils.get_incremental_state( - self, incremental_state, "cached_state" - ) - if cached_state is None: - return - - def reorder_state(state): - if isinstance(state, list): - return [reorder_state(state_i) for state_i in state] - return state.index_select(0, new_order) - - new_state = tuple(map(reorder_state, cached_state)) - utils.set_incremental_state(self, incremental_state, "cached_state", new_state) - - def max_positions(self): - """Maximum output length supported by the decoder.""" - return int(1e5) # an arbitrary large number - - -def Embedding(num_embeddings, embedding_dim, padding_idx): - m = nn.Embedding(num_embeddings, embedding_dim, padding_idx=padding_idx) - nn.init.uniform_(m.weight, -0.1, 0.1) - nn.init.constant_(m.weight[padding_idx], 0) - return m - - -def LSTM(input_size, hidden_size, **kwargs): - m = nn.LSTM(input_size, hidden_size, **kwargs) - for name, param in m.named_parameters(): - if "weight" in name or "bias" in name: - param.data.uniform_(-0.1, 0.1) - return m - - -def LSTMCell(input_size, hidden_size, **kwargs): - m = nn.LSTMCell(input_size, hidden_size, **kwargs) - for name, param in m.named_parameters(): - if "weight" in name or "bias" in name: - param.data.uniform_(-0.1, 0.1) - return m - - -def Linear(in_features, out_features, bias=True, dropout=0): - """Weight-normalized Linear layer (input: N x T x C)""" - m = nn.Linear(in_features, out_features, bias=bias) - m.weight.data.uniform_(-0.1, 0.1) - if bias: - m.bias.data.uniform_(-0.1, 0.1) - return m - - -@register_model_architecture("laser_lstm", "laser_lstm") -def base_architecture(args): - args.encoder_embed_dim = getattr(args, "encoder_embed_dim", 512) - args.encoder_embed_path = getattr(args, "encoder_embed_path", None) - args.encoder_hidden_size = getattr( - args, "encoder_hidden_size", args.encoder_embed_dim - ) - args.encoder_layers = getattr(args, "encoder_layers", 1) - args.encoder_bidirectional = getattr(args, "encoder_bidirectional", False) - args.encoder_dropout_in = getattr(args, "encoder_dropout_in", args.dropout) - args.encoder_dropout_out = getattr(args, "encoder_dropout_out", args.dropout) - args.decoder_embed_dim = getattr(args, "decoder_embed_dim", 512) - args.decoder_embed_path = getattr(args, "decoder_embed_path", None) - args.decoder_hidden_size = getattr( - args, "decoder_hidden_size", args.decoder_embed_dim - ) - args.decoder_layers = getattr(args, "decoder_layers", 1) - args.decoder_out_embed_dim = getattr(args, "decoder_out_embed_dim", 512) - args.decoder_dropout_in = getattr(args, "decoder_dropout_in", args.dropout) - args.decoder_dropout_out = getattr(args, "decoder_dropout_out", args.dropout) - args.decoder_zero_init = getattr(args, "decoder_zero_init", "0") - args.decoder_lang_embed_dim = getattr(args, "decoder_lang_embed_dim", 0) - args.fixed_embeddings = getattr(args, "fixed_embeddings", False) diff --git a/fairseq/examples/laser/laser_src/laser_task.py b/fairseq/examples/laser/laser_src/laser_task.py deleted file mode 100644 index e4152fde6861488acc3595fa25c456bf60f134b9..0000000000000000000000000000000000000000 --- a/fairseq/examples/laser/laser_src/laser_task.py +++ /dev/null @@ -1,331 +0,0 @@ -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. - - -from collections import OrderedDict, defaultdict -import json -import os -import logging -from argparse import ArgumentError - -from fairseq import options, models -from fairseq.data import ( - data_utils, - Dictionary, - LanguagePairDataset, - IndexedDataset, - FairseqDataset, -) -from .multitask_data_utils import ( - MultitaskDatasetWrapper, - MultidatasetEpochBatchIterator, -) - - -from fairseq.tasks import LegacyFairseqTask, register_task - -logger = logging.getLogger(__name__) - - -@register_task("laser") -class LaserTask(LegacyFairseqTask): - @staticmethod - def add_args(parser): - """Add task-specific arguments to the parser.""" - parser.add_argument( - "configfile", metavar="PATH", help="dataset configuration file in json" - ) - parser.add_argument( - "--weighting-alpha", - type=float, - default=None, - help="alpha for automatic weighting", - ) - parser.add_argument( - "--raw-text", action="store_true", help="load raw text dataset" - ) - parser.add_argument( - "--left-pad-source", - default="True", - type=str, - metavar="BOOL", - help="pad the source on the left (default: True)", - ) - parser.add_argument( - "--left-pad-target", - default="False", - type=str, - metavar="BOOL", - help="pad the target on the left (default: False)", - ) - try: - parser.add_argument( - "--max-source-positions", - default=1024, - type=int, - metavar="N", - help="max number of tokens in the source sequence", - ) - parser.add_argument( - "--max-target-positions", - default=1024, - type=int, - metavar="N", - help="max number of tokens in the target sequence", - ) - except ArgumentError: - # this might have already been defined. Once we transition this to hydra it should be fine to add it here. - pass - - def __init__(self, args, config, src_dictionary, tgt_dictionary, num_tasks): - super().__init__(args) - self.config = config - self.src_dictionary = src_dictionary - self.tgt_dictionary = tgt_dictionary - self.num_tasks = num_tasks - - @classmethod - def setup_task(cls, args, **kwargs): - with open(args.configfile, "r") as f: - config = json.load(f) - num_tasks = max(dataset["id"] for dataset in config["train"]) + 1 - - args.left_pad_source = options.eval_bool(args.left_pad_source) - args.left_pad_target = options.eval_bool(args.left_pad_target) - - src_dictionary = Dictionary.load(config["src_vocab"]) - tgt_dictionary = Dictionary.load(config["tgt_vocab"]) - - logger.info( - "| src Dictionary {} : {} types".format( - config["src_vocab"], len(src_dictionary) - ) - ) - logger.info( - "| tgt Dictionary {} : {} types".format( - config["tgt_vocab"], len(tgt_dictionary) - ) - ) - - return cls(args, config, src_dictionary, tgt_dictionary, num_tasks) - - # Experimental overriding for backtranslation - def build_model(self, args): - model = models.build_model(args, self) - return model - - def dataset(self, split): - if split not in self.datasets: - raise KeyError("Dataset not loaded: " + split) - return self.datasets[split] - - def load_dataset(self, split, epoch=1, **kwargs): - """Load a dataset split.""" - - def indexed_dataset(path, dictionary): - if self.args.raw_text: - raise Exception("Unable to handle raw text.") - dataset = IndexedDataset(path, fix_lua_indexing=True) - - return dataset - - pair_datasets = OrderedDict() - - if split == "valid": - self.datasets[split] = pair_datasets - return - - if split not in self.config: - raise FileNotFoundError( - "Dataset not found in config file: {}".format(split) - ) - - size_by_corpus = defaultdict(int) - size_sum = 0 - size_sum_with_subsampling = 0 - init_pair_datasets = {} - - for dataset_config in self.config[split]: - src_path = os.path.dirname(dataset_config["src"]) - corpus_name = src_path.split("/")[-2] - language_pair_name = src_path.split("/")[-1] - pair_datasets_key = corpus_name + "-" + language_pair_name - - logger.info(f"loading... {pair_datasets_key}") - if "src" in dataset_config: - src_dataset = indexed_dataset( - dataset_config["src"], self.src_dictionary - ) - else: - src_dataset = None - - if "tgt" in dataset_config: - tgt_dataset = indexed_dataset( - dataset_config["tgt"], self.tgt_dictionary - ) - else: - tgt_dataset = None - - dataset = LanguagePairDataset( - src_dataset, - src_dataset.sizes, - self.src_dictionary, - tgt_dataset, - tgt_dataset.sizes, - self.tgt_dictionary, - left_pad_source=self.args.left_pad_source, - left_pad_target=self.args.left_pad_target, - ) - - if pair_datasets_key in init_pair_datasets: - logger.warning( - f"Ignoring already added {pair_datasets_key}. " - f"Consider using `sample` key in order to upsample." - ) - else: - init_pair_datasets[pair_datasets_key] = { - "dataset": dataset, - "sample": dataset_config.get("sample", None), - "id": dataset_config.get("id", None), - "len": len(dataset), - } - - length_sum = 0 - weighted_freqs_sum = 0 - freq_per_dataset = {} - vmax = 0 - vmin = 1 - weighted_freq_per_dataset = {} - - if self.args.weighting_alpha: - for key in init_pair_datasets: - if init_pair_datasets[key]["sample"] is None: - length_sum += len(init_pair_datasets[key]["dataset"]) - - for key in init_pair_datasets: - if init_pair_datasets[key]["sample"] is None: - val = float(init_pair_datasets[key]["len"]) / length_sum - freq_per_dataset[key] = val - weighted_freqs_sum += val ** self.args.weighting_alpha - - for key in freq_per_dataset: - val = ( - freq_per_dataset[key] ** self.args.weighting_alpha - / weighted_freqs_sum - ) - vmin = min(vmin, val) - vmax = max(vmax, val) - weighted_freq_per_dataset[key] = val - - for pair_datasets_key in init_pair_datasets: - dataset_config = init_pair_datasets[pair_datasets_key] - dataset = dataset_config["dataset"] - sample = dataset_config["sample"] - if sample is None: - sample = 1.0 - - if pair_datasets_key in weighted_freq_per_dataset: - w = vmax / weighted_freq_per_dataset[pair_datasets_key] - sample = w - - sample = round(sample) - - initial_sample = sample - initial_pair_datasets_key = pair_datasets_key - - while sample >= 1.0: - assert ( - pair_datasets_key not in pair_datasets - ), f"{pair_datasets_key} already in" - size_sum_with_subsampling += len(dataset) - pair_datasets[pair_datasets_key] = MultitaskDatasetWrapper( - dataset, dataset_config.get("id", 0), 1.0, name=pair_datasets_key - ) - size_sum += len(dataset) - sample -= 1.0 - pair_datasets_key += "-up" - - assert sample < 1e-6, f"sample remains > 0 {pair_datasets_key}" - - logger.info( - f"added pair {initial_pair_datasets_key} length {len(dataset)} new_length = {len(dataset)*initial_sample}" - ) - size_by_corpus[corpus_name] += len(dataset) - - self.datasets[split] = pair_datasets - logger.info( - f"Datasets number = {len(self.datasets[split])} size = {size_sum} size_sum_with_subsampling = {size_sum_with_subsampling}" - ) - - @property - def source_dictionary(self): - return self.src_dictionary - - @property - def target_dictionary(self): - return self.tgt_dictionary - - def get_batch_iterator( - self, - dataset, - max_tokens=None, - max_sentences=None, - max_positions=None, - ignore_invalid_inputs=False, - required_batch_size_multiple=1, - seed=1, - num_shards=1, - shard_id=0, - num_workers=0, - epoch=1, - data_buffer_size=0, - disable_iterator_cache=False, - ): - - assert isinstance(dataset, OrderedDict) - assert len(dataset) - assert isinstance(dataset[next(iter(dataset))], FairseqDataset) - - # initialize the dataset with the correct starting epoch - for _, dt in dataset.items(): - dt.set_epoch(epoch) - - indices = OrderedDict() - batch_sampler = OrderedDict() - - with data_utils.numpy_seed(seed + epoch): - for key, dt in dataset.items(): - logger.info(f"\t ordered_indices {key}") - indices[key] = dt.ordered_indices() - - # filter examples that are too large - if max_positions is not None: - for key, dt in dataset.items(): - logger.info(f"\t filter_by_size {key}") - indices[key], ignored = dt.filter_indices_by_size( - indices[key], max_positions - ) - - for key, dt in dataset.items(): - logger.info(f"\t batch_by_size {key}") - batch_sampler[key] = data_utils.batch_by_size( - indices[key], - dt.num_tokens, - max_tokens=max_tokens, - max_sentences=max_sentences, - required_batch_size_multiple=required_batch_size_multiple, - ) - - epoch_iter = MultidatasetEpochBatchIterator( - dataset=dataset, - batch_sampler=batch_sampler, - seed=seed, - num_shards=num_shards, - shard_id=shard_id, - num_workers=num_workers, - epoch=epoch, - ) - - return epoch_iter diff --git a/fairseq/examples/laser/laser_src/laser_transformer.py b/fairseq/examples/laser/laser_src/laser_transformer.py deleted file mode 100644 index 0be030994ff87334ca0392302374693f7f2c61b3..0000000000000000000000000000000000000000 --- a/fairseq/examples/laser/laser_src/laser_transformer.py +++ /dev/null @@ -1,354 +0,0 @@ -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. - -import logging - -from typing import Any, Dict, List, Optional -from torch import Tensor - -import torch -import torch.nn as nn - -from fairseq.models import ( - FairseqEncoderDecoderModel, - register_model, - register_model_architecture, -) -from fairseq.models.transformer import ( - base_architecture, - Embedding, - TransformerModel, - TransformerEncoder, - TransformerDecoder, -) -from fairseq.modules import ( - TransformerDecoderLayer, -) - -logger = logging.getLogger(__name__) - - -@register_model("laser_transformer") -class LaserTransformerModel(FairseqEncoderDecoderModel): - """Train Transformer for LASER task - - Requires --task laser - """ - - def __init__(self, encoder, decoder): - super().__init__(encoder, decoder) - - def forward( - self, - src_tokens, - src_lengths, - prev_output_tokens=None, - tgt_tokens=None, - tgt_lengths=None, - target_language_id=-1, - dataset_name="", - ): - laser_encoder_out = self.encoder(src_tokens, src_lengths) - return self.decoder( - prev_output_tokens, laser_encoder_out, lang_id=target_language_id - ) - - @staticmethod - def add_args(parser): - """Add model-specific arguments to the parser.""" - TransformerModel.add_args(parser) - parser.add_argument( - "--decoder-lang-embed-dim", - type=int, - metavar="N", - help="decoder language embedding dimension", - ) - - @classmethod - def build_model(cls, args, task): - base_laser_transformer_architecture(args) - - num_langs = task.num_tasks if hasattr(task, "num_tasks") else 0 - - def load_embed_tokens(dictionary, embed_dim): - num_embeddings = len(dictionary) - padding_idx = dictionary.pad() - - return Embedding(num_embeddings, embed_dim, padding_idx) - - encoder_embed_tokens = load_embed_tokens( - task.source_dictionary, args.encoder_embed_dim - ) - decoder_embed_tokens = load_embed_tokens( - task.target_dictionary, args.decoder_embed_dim - ) - num_langs = task.num_tasks if hasattr(task, "num_tasks") else 0 - - encoder = LaserTransformerEncoder( - args, task.source_dictionary, encoder_embed_tokens - ) - - decoder = LaserTransformerDecoder( - args, - task.target_dictionary, - decoder_embed_tokens, - num_langs=num_langs, - lang_embed_dim=args.decoder_lang_embed_dim, - ) - - return cls(encoder, decoder) - - -class LaserTransformerEncoder(TransformerEncoder): - def __init__(self, *args, **kwargs): - super().__init__(*args, **kwargs) - - def forward(self, src_tokens, *args, **kwargs): - encoder_out = super().forward(src_tokens, *args, **kwargs) - - x = encoder_out["encoder_out"][0] # T x B x C - padding_mask = src_tokens.eq(self.padding_idx).t().unsqueeze(-1) - - if padding_mask.any(): - x = x.float().masked_fill_(padding_mask, float("-inf")).type_as(x) - - # Build the sentence embedding by max-pooling over the encoder outputs - sentemb = x.max(dim=0)[0] - - # The Pytorch Mobile lite interpreter does not supports returning NamedTuple in - # `foward` so we use a dictionary instead. - # TorchScript does not support mixed values so the values are all lists. - # The empty list is equivalent to None. - return {"sentemb": [sentemb]} # B x C - - @torch.jit.export - def reorder_encoder_out(self, encoder_out: Dict[str, List[Tensor]], new_order): - """ - Same as the one in transformer.py, with new_sentemb - """ - if len(encoder_out["sentemb"]) == 0: - new_sentemb = [] - else: - new_sentemb = [encoder_out["sentemb"][0].index_select(0, new_order)] - - return { - "sentemb": new_sentemb, # B x C - } - - -class LaserTransformerDecoder(TransformerDecoder): - def __init__(self, args, dictionary, *kargs, **kwargs): - self.num_langs = kwargs.get("num_langs", 1) - self.lang_embed_dim = kwargs.get("lang_embed_dim", 0) - kwargs.pop("num_langs", None) - kwargs.pop("lang_embed_dim", None) - - super().__init__(args, dictionary, *kargs, **kwargs, no_encoder_attn=True) - - if self.lang_embed_dim == 0: - self.embed_lang = None - else: - self.embed_lang = nn.Embedding(self.num_langs, self.lang_embed_dim) - nn.init.uniform_(self.embed_lang.weight, -0.1, 0.1) - - if self.output_projection is not None: - laser_output_embed_dim = ( - self.output_embed_dim + self.lang_embed_dim + args.encoder_embed_dim - ) - self.output_projection = nn.Linear( - laser_output_embed_dim, len(dictionary), bias=False - ) - nn.init.normal_( - self.output_projection.weight, - mean=0, - std=laser_output_embed_dim ** -0.5, - ) - - def build_decoder_layer(self, args, no_encoder_attn=False): - decoder_embed_dim = args.decoder_embed_dim - args.decoder_embed_dim = ( - decoder_embed_dim + self.lang_embed_dim + args.encoder_embed_dim - ) - res = TransformerDecoderLayer(args, no_encoder_attn=True) - args.decoder_embed_dim = decoder_embed_dim - - return res - - def extract_features( - self, - prev_output_tokens, - encoder_out: Optional[Dict[str, List[Tensor]]], - incremental_state: Optional[Dict[str, Dict[str, Optional[Tensor]]]] = None, - full_context_alignment: bool = False, - alignment_layer: Optional[int] = None, - alignment_heads: Optional[int] = None, - lang_id: Optional[int] = None, - ): - """ - Similar to *forward* but only return features. - - Includes several features from "Jointly Learning to Align and - Translate with Transformer Models" (Garg et al., EMNLP 2019). - - Args: - full_context_alignment (bool, optional): don't apply - auto-regressive mask to self-attention (default: False). - alignment_layer (int, optional): return mean alignment over - heads at this layer (default: last layer). - alignment_heads (int, optional): only average alignment over - this many heads (default: all heads). - - Returns: - tuple: - - the decoder's features of shape `(batch, tgt_len, embed_dim)` - - a dictionary with any model-specific outputs - """ - if alignment_layer is None: - alignment_layer = self.num_layers - 1 - - # embed positions - positions = ( - self.embed_positions( - prev_output_tokens, incremental_state=incremental_state - ) - if self.embed_positions is not None - else None - ) - - if incremental_state is not None: - prev_output_tokens = prev_output_tokens[:, -1:] - if positions is not None: - positions = positions[:, -1:] - - bsz, seqlen = prev_output_tokens.size() - - # embed tokens and positions - x = self.embed_scale * self.embed_tokens(prev_output_tokens) - - if self.quant_noise is not None: - x = self.quant_noise(x) - - if self.project_in_dim is not None: - x = self.project_in_dim(x) - - if positions is not None: - x += positions - - if self.layernorm_embedding is not None: - x = self.layernorm_embedding(x) - - x = self.dropout_module(x) - - # B x T x C -> T x B x C - x = x.transpose(0, 1) - - if self.embed_lang is not None: - lang_ids = prev_output_tokens.data.new_full((bsz,), lang_id) - langemb = self.embed_lang(lang_ids) - langemb = langemb.unsqueeze(0) - repeat_vals = [x.shape[0] // langemb.shape[0]] + [-1] * ( - len(langemb.shape) - 1 - ) - x = torch.cat((x, langemb.expand(*repeat_vals)), dim=-1) - - sentemb = encoder_out["sentemb"][0] - sentemb = sentemb.unsqueeze(0) - - repeat_vals = [x.shape[0] // sentemb.shape[0]] + [-1] * (len(sentemb.shape) - 1) - x = torch.cat((x, sentemb.expand(*repeat_vals)), dim=-1) - - self_attn_padding_mask: Optional[Tensor] = None - if self.cross_self_attention or prev_output_tokens.eq(self.padding_idx).any(): - self_attn_padding_mask = prev_output_tokens.eq(self.padding_idx) - - # decoder layers - attn: Optional[Tensor] = None - inner_states: List[Optional[Tensor]] = [x] - for idx, layer in enumerate(self.layers): - if incremental_state is None and not full_context_alignment: - self_attn_mask = self.buffered_future_mask(x) - else: - self_attn_mask = None - - x, layer_attn, _ = layer( - x, - None, - None, - incremental_state, - self_attn_mask=self_attn_mask, - self_attn_padding_mask=self_attn_padding_mask, - need_attn=bool((idx == alignment_layer)), - need_head_weights=bool((idx == alignment_layer)), - ) - inner_states.append(x) - if layer_attn is not None and idx == alignment_layer: - attn = layer_attn.float().to(x) - - if attn is not None: - if alignment_heads is not None: - attn = attn[:alignment_heads] - - # average probabilities over heads - attn = attn.mean(dim=0) - - if self.layer_norm is not None: - x = self.layer_norm(x) - - # T x B x C -> B x T x C - x = x.transpose(0, 1) - - if self.project_out_dim is not None: - x = self.project_out_dim(x) - - return x, {"attn": [attn], "inner_states": inner_states} - - def forward( - self, - prev_output_tokens, - encoder_out: Optional[Dict[str, List[Tensor]]] = None, - incremental_state: Optional[Dict[str, Dict[str, Optional[Tensor]]]] = None, - features_only: bool = False, - alignment_layer: Optional[int] = None, - alignment_heads: Optional[int] = None, - src_lengths: Optional[Any] = None, - return_all_hiddens: bool = False, - lang_id: Optional[int] = None, - ): - """ - Args: - prev_output_tokens (LongTensor): previous decoder outputs of shape - `(batch, tgt_len)`, for teacher forcing - encoder_out (optional): output from the encoder, used for - encoder-side attention - incremental_state (dict): dictionary used for storing state during - :ref:`Incremental decoding` - features_only (bool, optional): only return features without - applying output layer (default: False). - - Returns: - tuple: - - the decoder's output of shape `(batch, tgt_len, vocab)` - - a dictionary with any model-specific outputs - """ - - assert lang_id is not None - - x, extra = self.extract_features( - prev_output_tokens, - encoder_out=encoder_out, - incremental_state=incremental_state, - alignment_layer=alignment_layer, - alignment_heads=alignment_heads, - lang_id=lang_id, - ) - if not features_only: - x = self.output_layer(x) - return x, extra - - -@register_model_architecture("laser_transformer", "laser_transformer") -def base_laser_transformer_architecture(args): - base_architecture(args) - args.decoder_lang_embed_dim = getattr(args, "decoder_lang_embed_dim", 0) diff --git a/fairseq/examples/laser/laser_src/multitask_data_utils.py b/fairseq/examples/laser/laser_src/multitask_data_utils.py deleted file mode 100644 index b05caea26793bf5112a7abc29d76225f578f3ebe..0000000000000000000000000000000000000000 --- a/fairseq/examples/laser/laser_src/multitask_data_utils.py +++ /dev/null @@ -1,143 +0,0 @@ -# Copyright (c) Facebook, Inc. and its affiliates. -# -# This source code is licensed under the MIT license found in the -# LICENSE file in the root directory of this source tree. - -from collections import OrderedDict - -import numpy as np - -from fairseq.data import BaseWrapperDataset, FairseqDataset, iterators - - -class MultiItr(object): - def __init__(self, itr): - self.itr = itr - self._counts = [0 for x in itr] - - def __len__(self): - return sum(len(itr) for itr in self.itr) - - def __iter__(self): - return self - - def __next__(self): - ratios = [count / len(itr) for count, itr in zip(self._counts, self.itr)] - idx = ratios.index(min(ratios)) - self._counts[idx] += 1 - return next(self.itr[idx]) - - -class MultidatasetEpochBatchIterator(iterators.EpochBatchIterating): - """A wrapper around multiple epoch batch iterators.""" - - def __init__( - self, - dataset, - batch_sampler, - seed=1, - num_shards=1, - shard_id=0, - num_workers=0, - epoch=1, - ): - - assert isinstance(dataset, OrderedDict) - assert len(dataset) - assert isinstance(dataset[next(iter(dataset))], FairseqDataset) - - self.iterators = [] - - self.epoch = epoch - for key, dt in dataset.items(): - epoch_iter = iterators.EpochBatchIterator( - dataset=dt, - collate_fn=dt.collater, - batch_sampler=batch_sampler[key], - seed=seed, - num_shards=num_shards, - shard_id=shard_id, - num_workers=0, - epoch=epoch, - ) - self.iterators.append(epoch_iter) - - def __len__(self): - return sum(len(itr) for itr in self.iterators) - - def next_epoch_itr(self, shuffle=True, fix_batches_to_gpus=False): - # `self.epoch += 1` should be handled by underlying `EpochBatchIterator`s. - return MultiItr( - [ - itr.next_epoch_itr( - shuffle=shuffle, fix_batches_to_gpus=fix_batches_to_gpus - ) - for itr in self.iterators - ] - ) - - def end_of_epoch(self): - return all(itr.end_of_epoch() for itr in self.iterators) - - @property - def next_epoch_idx(self): - """Return the epoch index after *next_epoch_itr* is called.""" - - epochs = [itr.next_epoch_idx for itr in self.iterators] - self.epoch = epochs[0] - assert all(epoch == self.epoch for epoch in epochs) - - return self.epoch - - @property - def iterations_in_epoch(self): - return sum(itr.iterations_in_epoch for itr in self.iterators) - - def state_dict(self): - return { - "iterators": [it.state_dict() for it in self.iterators], - "epoch": self.epoch, - } - - def load_state_dict(self, state_dict): - self.epoch = state_dict["epoch"] - for it, d in zip(self.iterators, state_dict["iterators"]): - it.load_state_dict(d) - - -class MultitaskDatasetWrapper(BaseWrapperDataset): - """A wrapper for a multitask dataset.""" - - def __init__(self, dataset, target_language_id, sample=1.0, name=""): - super().__init__(dataset) - self.target_language_id = target_language_id - self.sample = sample - self.name = name - - def collater(self, *args, **kwargs): - ans = self.dataset.collater(*args, **kwargs) - if "net_input" in ans: - ans["net_input"]["target_language_id"] = self.target_language_id - ans["net_input"]["dataset_name"] = self.name - return ans - - def num_tokens(self, *args, **kwargs): - return self.dataset.num_tokens(*args, **kwargs) - - def ordered_indices(self, *args, **kwargs): - indices = self.dataset.ordered_indices(*args, **kwargs) - # Hacky solution for sampling - size = int(self.sample * indices.shape[0]) - - return indices.take(np.sort(np.random.permutation(indices.shape[0])[:size])) - - def size(self, index: int): - return self.dataset.size(index) - - @property - def supports_prefetch(self): - """Whether this dataset supports prefetching.""" - return getattr(self.dataset, "supports_prefetch", False) - - def prefetch(self, indices): - return self.dataset.prefetch(indices) diff --git a/fairseq/examples/latent_depth/README.md b/fairseq/examples/latent_depth/README.md deleted file mode 100644 index 7774c333053b95d15b180fdfc3ee3cd817790520..0000000000000000000000000000000000000000 --- a/fairseq/examples/latent_depth/README.md +++ /dev/null @@ -1,77 +0,0 @@ -# Deep Transformers with Latent Depth (Li et al., 2020) - -[https://arxiv.org/abs/2009.13102](https://arxiv.org/abs/2009.13102). - -## Introduction - -We present a probabilistic framework to automatically learn which layer(s) to use by learning the posterior distributions of layer selection. As an extension of this framework, we propose a novel method to train one shared Transformer network for multilingual machine translation with different layer selection posteriors for each language pair. - -## Training a multilingual model with latent depth - -Below is an example of training with latent depth in decoder for one-to-many (O2M) related languages. We use the same preprocessed (numberized and binarized) TED8 dataset as in [Balancing Training for Multilingual Neural Machine Translation (Wang et al., 2020)](https://github.com/cindyxinyiwang/multiDDS), which could be generated by [the script](https://github.com/cindyxinyiwang/multiDDS/blob/multiDDS/util_scripts/prepare_multilingual_data.sh) the author provided. -```bash -lang_pairs_str="eng-aze,eng-bel,eng-ces,eng-glg,eng-por,eng-rus,eng-slk,eng-tur" -databin_dir=