Update model_args.json

model_args.json

@@ -1 +1 @@
-{"doc_stride": 512, "max_query_length": 64, "n_best_size": 20, "max_answer_length": 100, "null_score_diff_threshold": 0.0, "adam_epsilon": 1e-08, "best_model_dir": "outputs/best_model", "cache_dir": "cache_dir/", "config": {}, "do_lower_case": false, "early_stopping_consider_epochs": false, "early_stopping_delta": 0, "early_stopping_metric": "correct", "early_stopping_metric_minimize": false, "early_stopping_patience": 3, "encoding": null, "eval_batch_size": 16, "evaluate_during_training": false, "evaluate_during_training_silent": true, "evaluate_during_training_steps": 2000, "evaluate_during_training_verbose": false, "fp16": true, "fp16_opt_level": "O2", "gradient_accumulation_steps": 4, "learning_rate": 3e-06, "local_rank": -1, "logging_steps": 50, "manual_seed": null, "max_grad_norm": 1.0, "max_seq_length": 896, "multiprocessing_chunksize": 500, "n_gpu": 1, "no_cache": true, "no_save": false, "num_train_epochs": 4, "output_dir": "longformer-squadv2", "overwrite_output_dir": true, "process_count": 22, "reprocess_input_data": false, "save_best_model": true, "save_eval_checkpoints": true, "save_model_every_epoch": false, "save_steps": 2000, "save_optimizer_and_scheduler": true, "silent": false, "tensorboard_dir": null, "train_batch_size": 4, "use_cached_eval_features": false, "use_early_stopping": false, "use_multiprocessing": true, "wandb_kwargs": {}, "wandb_project": null, "warmup_ratio": 0.06, "warmup_steps": 1956, "weight_decay": 0, "model_name": "mrm8488/longformer-base-4096-finetuned-squadv2", "model_type": "longformer"}
+{"adam_epsilon": 1e-08, "best_model_dir": "outputs/best_model", "cache_dir": "cache_dir/", "config": {}, "do_lower_case": false, "early_stopping_consider_epochs": false, "early_stopping_delta": 0, "early_stopping_metric": "correct", "early_stopping_metric_minimize": false, "early_stopping_patience": 3, "encoding": null, "eval_batch_size": 80, "evaluate_during_training": false, "evaluate_during_training_silent": true, "evaluate_during_training_steps": 2000, "evaluate_during_training_verbose": false, "fp16": true, "fp16_opt_level": "O2", "gradient_accumulation_steps": 1, "learning_rate": 1e-06, "local_rank": -1, "logging_steps": 1, "manual_seed": null, "max_grad_norm": 1.0, "max_seq_length": 512, "multiprocessing_chunksize": 500, "n_gpu": 1, "no_cache": true, "no_save": false, "num_train_epochs": 15, "output_dir": "longformer-squadv2", "overwrite_output_dir": true, "process_count": 22, "reprocess_input_data": false, "save_best_model": true, "save_eval_checkpoints": true, "save_model_every_epoch": false, "save_steps": 2000, "save_optimizer_and_scheduler": true, "silent": false, "tensorboard_dir": null, "train_batch_size": 20, "use_cached_eval_features": false, "use_early_stopping": false, "use_multiprocessing": true, "wandb_kwargs": {}, "wandb_project": null, "warmup_ratio": 0.06, "warmup_steps": 11314, "weight_decay": 0, "doc_stride": 512, "max_query_length": 64, "n_best_size": 20, "max_answer_length": 100, "null_score_diff_threshold": 0.0}