Update README.md
Browse files
README.md
CHANGED
@@ -5,18 +5,16 @@ widget:
|
|
5 |
---
|
6 |
# wangchanberta-base-att-spm-uncased-finetune-qa
|
7 |
|
8 |
-
Finetuning `wangchanberta-base-att-spm-uncased` with the training set of `iapp_wiki_qa_squad
|
9 |
-
|
10 |
-
|
11 |
```
|
12 |
-
export
|
13 |
-
|
14 |
-
export MODEL_NAME=wangchanberta-base-att-spm-uncased
|
15 |
python train_question_answering_lm_finetuning.py \
|
16 |
--model_name $MODEL_NAME \
|
17 |
-
--dataset_name
|
18 |
-
--output_dir $MODEL_NAME-finetune-
|
19 |
-
--log_dir $MODEL_NAME-finetune-
|
20 |
--lowercase \
|
21 |
--pad_on_right \
|
22 |
--fp16
|
|
|
5 |
---
|
6 |
# wangchanberta-base-att-spm-uncased-finetune-qa
|
7 |
|
8 |
+
Finetuning `airesearch/wangchanberta-base-att-spm-uncased` with the training set of `iapp_wiki_qa_squad`, `thaiqa`, `xquad` (removed examples which have cosine similarity with validation and test examples over 0.8). Benchmarks shared on [wandb](https://wandb.ai/cstorm125/wangchanberta-qa) using validation and test sets of `iapp_wiki_qa_squad`.
|
9 |
+
Trained with [thai2transformers](https://github.com/vistec-AI/thai2transformers/blob/dev/scripts/downstream/train_question_answering_lm_finetuning.py). Combined benchmark datasets downloadable at [vistec-ai/thai2transformers](https://github.com/vistec-AI/thai2transformers/releases/tag/qa-v0.2).
|
10 |
+
Run with:
|
11 |
```
|
12 |
+
export MODEL_NAME=airesearch/wangchanberta-base-att-spm-uncased
|
|
|
|
|
13 |
python train_question_answering_lm_finetuning.py \
|
14 |
--model_name $MODEL_NAME \
|
15 |
+
--dataset_name iapp_thaiqa_xquad \
|
16 |
+
--output_dir $MODEL_NAME-finetune-iapp_thaiqa_xquad-model \
|
17 |
+
--log_dir $MODEL_NAME-finetune-iapp_thaiqa_xquad-log \
|
18 |
--lowercase \
|
19 |
--pad_on_right \
|
20 |
--fp16
|