mirror of
https://github.com/huggingface/transformers.git
synced 2025-07-28 08:42:23 +06:00
54f9fbeff8
* Fully rework training/prediction loops * fix method name * Fix variable name * Fix property name * Fix scope * Fix method name * Fix tuple index * Fix tuple index * Fix indentation * Fix variable name * fix eval before log * Add drop remainder for test dataset * Fix step number + fix logging datetime * fix eval loss value * use global step instead of step + fix logging at step 0 * Fix logging datetime * Fix global_step usage * Fix breaking loop + logging datetime * Fix step in prediction loop * Fix step breaking * Fix train/test loops * Force TF at least 2.2 for the trainer * Use assert_cardinality to facilitate the dataset size computation * Log steps per epoch * Make tfds compliant with TPU * Make tfds compliant with TPU * Use TF dataset enumerate instead of the Python one * revert previous commit * Fix data_dir * Apply style * rebase on master * Address Sylvain's comments * Address Sylvain's and Lysandre comments * Trigger CI * Remove unused import |
||
|---|---|---|
| .. | ||
| README.md | ||
| run_multiple_choice.py | ||
| run_tf_multiple_choice.py | ||
| utils_multiple_choice.py | ||
Multiple Choice
Based on the script run_multiple_choice.py.
Fine-tuning on SWAG
Download swag data
#training on 4 tesla V100(16GB) GPUS
export SWAG_DIR=/path/to/swag_data_dir
python ./examples/multiple-choice/run_multiple_choice.py \
--task_name swag \
--model_name_or_path roberta-base \
--do_train \
--do_eval \
--data_dir $SWAG_DIR \
--learning_rate 5e-5 \
--num_train_epochs 3 \
--max_seq_length 80 \
--output_dir models_bert/swag_base \
--per_gpu_eval_batch_size=16 \
--per_device_train_batch_size=16 \
--gradient_accumulation_steps 2 \
--overwrite_output
Training with the defined hyper-parameters yields the following results:
***** Eval results *****
eval_acc = 0.8338998300509847
eval_loss = 0.44457291918821606
Tensorflow
export SWAG_DIR=/path/to/swag_data_dir
python ./examples/multiple-choice/run_tf_multiple_choice.py \
--task_name swag \
--model_name_or_path bert-base-cased \
--do_train \
--do_eval \
--data_dir $SWAG_DIR \
--learning_rate 5e-5 \
--num_train_epochs 3 \
--max_seq_length 80 \
--output_dir models_bert/swag_base \
--per_gpu_eval_batch_size=16 \
--per_device_train_batch_size=16 \
--logging-dir logs \
--gradient_accumulation_steps 2 \
--overwrite_output
Run it in colab
