| { | |
| "best_metric": 73.71718882303134, | |
| "best_model_checkpoint": "./whisper-tiny-hindi2_test/checkpoint-120", | |
| "epoch": 4.920634920634921, | |
| "eval_steps": 40, | |
| "global_step": 155, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6349206349206349, | |
| "grad_norm": Infinity, | |
| "learning_rate": 2.043269230769231e-06, | |
| "loss": 3.3856, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.2698412698412698, | |
| "grad_norm": 12.90635871887207, | |
| "learning_rate": 4.447115384615384e-06, | |
| "loss": 2.2169, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.2698412698412698, | |
| "eval_loss": 1.7869399785995483, | |
| "eval_runtime": 509.7564, | |
| "eval_samples_per_second": 1.962, | |
| "eval_steps_per_second": 0.49, | |
| "eval_wer": 241.5580016934801, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 8.530549049377441, | |
| "learning_rate": 6.850961538461537e-06, | |
| "loss": 1.4435, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.5396825396825395, | |
| "grad_norm": 7.451807022094727, | |
| "learning_rate": 9.254807692307693e-06, | |
| "loss": 1.0037, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.5396825396825395, | |
| "eval_loss": 0.8910112380981445, | |
| "eval_runtime": 318.2676, | |
| "eval_samples_per_second": 3.142, | |
| "eval_steps_per_second": 0.786, | |
| "eval_wer": 78.55207451312447, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1746031746031744, | |
| "grad_norm": 6.283768177032471, | |
| "learning_rate": 1.1658653846153846e-05, | |
| "loss": 0.7501, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 6.983267307281494, | |
| "learning_rate": 1.40625e-05, | |
| "loss": 0.5907, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "eval_loss": 0.624940812587738, | |
| "eval_runtime": 318.0146, | |
| "eval_samples_per_second": 3.145, | |
| "eval_steps_per_second": 0.786, | |
| "eval_wer": 73.71718882303134, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 5.939335346221924, | |
| "learning_rate": 1.646634615384615e-05, | |
| "loss": 0.4806, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.920634920634921, | |
| "step": 155, | |
| "total_flos": 1.2132186587136e+17, | |
| "train_loss": 1.3161909626376245, | |
| "train_runtime": 1802.5104, | |
| "train_samples_per_second": 2.774, | |
| "train_steps_per_second": 0.086 | |
| }, | |
| { | |
| "epoch": 4.920634920634921, | |
| "step": 155, | |
| "total_flos": 1.2132186587136e+17, | |
| "train_loss": 0.0, | |
| "train_runtime": 0.1541, | |
| "train_samples_per_second": 32447.028, | |
| "train_steps_per_second": 1005.858 | |
| }, | |
| { | |
| "epoch": 4.920634920634921, | |
| "step": 155, | |
| "total_flos": 1.2132186587136e+17, | |
| "train_loss": 0.0, | |
| "train_runtime": 0.2109, | |
| "train_samples_per_second": 23707.402, | |
| "train_steps_per_second": 734.929 | |
| }, | |
| { | |
| "epoch": 4.920634920634921, | |
| "step": 155, | |
| "total_flos": 1.2132186587136e+17, | |
| "train_loss": 0.0, | |
| "train_runtime": 0.17, | |
| "train_samples_per_second": 29405.194, | |
| "train_steps_per_second": 911.561 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 155, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 40, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2132186587136e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |