Update README.md
Browse files
README.md
CHANGED
@@ -60,20 +60,21 @@ To reproduce results in Google Colab:
|
|
60 |
- Run this python code:
|
61 |
|
62 |
```python
|
63 |
-
python /content/transformers/examples/pytorch/question-answering/run_qa.py --model_name_or_path sultan/BioM-ELECTRA-Large-SQuAD2
|
64 |
-
--do_eval
|
65 |
-
--version_2_with_negative
|
66 |
-
--per_device_eval_batch_size 8
|
67 |
-
--dataset_name squad_v2
|
68 |
-
--overwrite_output_dir
|
69 |
-
--fp16
|
70 |
--output_dir out
|
71 |
```
|
72 |
|
73 |
-
You don't need to download the SQuAD2 dataset. The code will download it from the HuggingFace datasets hub.
|
74 |
|
|
|
75 |
|
76 |
-
|
77 |
|
78 |
# Acknowledgment
|
79 |
|
|
|
60 |
- Run this python code:
|
61 |
|
62 |
```python
|
63 |
+
python /content/transformers/examples/pytorch/question-answering/run_qa.py --model_name_or_path sultan/BioM-ELECTRA-Large-SQuAD2 \
|
64 |
+
--do_eval \
|
65 |
+
--version_2_with_negative \
|
66 |
+
--per_device_eval_batch_size 8 \
|
67 |
+
--dataset_name squad_v2 \
|
68 |
+
--overwrite_output_dir \
|
69 |
+
--fp16 \
|
70 |
--output_dir out
|
71 |
```
|
72 |
|
73 |
+
- You don't need to download the SQuAD2 dataset. The code will download it from the HuggingFace datasets hub.
|
74 |
|
75 |
+
- Check our GitHub repo at https://github.com/salrowili/BioM-Transformers for TensorFlow and GluonNLP checkpoints.
|
76 |
|
77 |
+
- We added examples to fine-tune BioM-ELECTRA-Large on SQuAD and BioASQ7B using TensorFlow and TPU here https://github.com/salrowili/BioM-Transformers/tree/main/examples . In this example we show that we achieve 88.22 score in SQuAD2.0 since Tensor Flow code has Layer-wise decay feature.
|
78 |
|
79 |
# Acknowledgment
|
80 |
|