clinical_longformer_squadv2_maxlen320

This model is a fine-tuned version of yikuan8/Clinical-Longformer on the squad_v2 dataset using a max_seq_length of 320.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Tuning script used:

set BASE_MODEL=yikuan8/Clinical-Longformer
set OUTPUT_DIR=U:\Documents...

python run_qa.py ^
  --model_name_or_path %BASE_MODEL% ^
  --dataset_name squad_v2 ^
  --do_train ^
  --do_eval ^
  --version_2_with_negative ^
  --per_device_train_batch_size 4 ^
  --per_device_eval_batch_size 4 ^
  --gradient_accumulation_steps 4 ^
  --learning_rate 2e-5 ^
  --num_train_epochs 3 ^
  --max_seq_length 320 ^
  --doc_stride 128 ^
  --weight_decay 0.01 ^
  --fp16 ^
  --output_dir %OUTPUT_DIR% ^
  --overwrite_output_dir

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3.0
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.48.0.dev0
• Pytorch 2.0.1+cu117
• Datasets 3.0.1
• Tokenizers 0.21.0