lhbonifacio commited on
Commit
b508e39
1 Parent(s): 003fe39

Initial commit

Browse files
README.md ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: pt
3
+ license: mit
4
+ tags:
5
+ - msmarco
6
+ - t5
7
+ - pytorch
8
+ - tensorflow
9
+ - en
10
+ datasets:
11
+ - msmarco
12
+ widget:
13
+ - text: "Texto de exemplo em português"
14
+ inference: false
15
+ ---
16
+ # mt5-base-en-msmarco Reranker finetuned on MS MARCO
17
+ ## Introduction
18
+ mT5-base-en-msmarco is a mT5-based model finetuned on English MS MARCO passage dataset.
19
+ Further information about the dataset or the translation method can be found on our paper [**mMARCO: A Multilingual Version of the MS MARCO Passage Ranking Dataset**](https://arxiv.org/abs/2108.13897) and [mMARCO](https://github.com/unicamp-dl/mMARCO) repository.
20
+
21
+ ## Usage
22
+ ```python
23
+
24
+ from transformers import T5Tokenizer, MT5ForConditionalGeneration
25
+
26
+ model_name = 'unicamp-dl/mt5-base-en-msmarco'
27
+ tokenizer = T5Tokenizer.from_pretrained(model_name)
28
+ model = MT5ForConditionalGeneration.from_pretrained(model_name)
29
+
30
+ ```
31
+ # Citation
32
+ If you use mT5-base-en-msmarco, please cite:
33
+
34
+ @misc{bonifacio2021mmarco,
35
+ title={mMARCO: A Multilingual Version of the MS MARCO Passage Ranking Dataset},
36
+ author={Luiz Henrique Bonifacio and Israel Campiotti and Vitor Jeronymo and Roberto Lotufo and Rodrigo Nogueira},
37
+ year={2021},
38
+ eprint={2108.13897},
39
+ archivePrefix={arXiv},
40
+ primaryClass={cs.CL}
41
+ }
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "mt5/",
3
+ "architectures": [
4
+ "MT5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 2048,
7
+ "d_kv": 64,
8
+ "d_model": 768,
9
+ "decoder_start_token_id": 0,
10
+ "dropout_rate": 0.1,
11
+ "eos_token_id": 1,
12
+ "feed_forward_proj": "gated-gelu",
13
+ "initializer_factor": 1.0,
14
+ "is_encoder_decoder": true,
15
+ "layer_norm_epsilon": 1e-06,
16
+ "model_type": "mt5",
17
+ "num_decoder_layers": 12,
18
+ "num_heads": 12,
19
+ "num_layers": 12,
20
+ "output_past": true,
21
+ "pad_token_id": 0,
22
+ "relative_attention_num_buckets": 32,
23
+ "tie_word_embeddings": false,
24
+ "tokenizer_class": "T5Tokenizer",
25
+ "torch_dtype": "float32",
26
+ "transformers_version": "4.11.3",
27
+ "use_cache": true,
28
+ "vocab_size": 250112
29
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46614554aa37abe590f8690722195becd4e7fafe10dbdbd326033c91f60a7533
3
+ size 2329696333
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef78f86560d809067d12bac6c09f19a462cb3af3f54d2b8acbba26e1433125d6
3
+ size 4309802
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 0, "additional_special_tokens": null, "sp_model_kwargs": {}, "special_tokens_map_file": "/home/patrick/.cache/torch/transformers/685ac0ca8568ec593a48b61b0a3c272beee9bc194a3c7241d15dcadb5f875e53.f76030f3ec1b96a8199b2593390c610e76ca8028ef3d24680000619ffb646276", "tokenizer_file": null, "name_or_path": "google/mt5-base", "tokenizer_class": "T5Tokenizer"}