| { | |
| "best_metric": 0.7213644980359202, | |
| "best_model_checkpoint": "/home/phison/LargeFiles/regional-asr/whisper-base-bn/checkpoint-264", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 264, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11363636363636363, | |
| "grad_norm": 39.91372299194336, | |
| "learning_rate": 1.4000000000000001e-06, | |
| "loss": 6.1886, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "grad_norm": 20.913997650146484, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "loss": 4.1637, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3409090909090909, | |
| "grad_norm": 10.190018653869629, | |
| "learning_rate": 5.400000000000001e-06, | |
| "loss": 2.967, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.45454545454545453, | |
| "grad_norm": 6.0802483558654785, | |
| "learning_rate": 7.4e-06, | |
| "loss": 2.1864, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5681818181818182, | |
| "grad_norm": 5.394363880157471, | |
| "learning_rate": 9.4e-06, | |
| "loss": 1.6717, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6818181818181818, | |
| "grad_norm": 3.06591534614563, | |
| "learning_rate": 9.672897196261683e-06, | |
| "loss": 1.3815, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7954545454545454, | |
| "grad_norm": 2.8467049598693848, | |
| "learning_rate": 9.205607476635515e-06, | |
| "loss": 1.2658, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 2.3901357650756836, | |
| "learning_rate": 8.738317757009348e-06, | |
| "loss": 1.1666, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.0611085891723633, | |
| "eval_runtime": 538.9962, | |
| "eval_samples_per_second": 5.165, | |
| "eval_steps_per_second": 0.161, | |
| "eval_wer": 0.7727432625222302, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.0227272727272727, | |
| "grad_norm": 2.7212612628936768, | |
| "learning_rate": 8.271028037383178e-06, | |
| "loss": 1.1132, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1363636363636362, | |
| "grad_norm": 2.4905169010162354, | |
| "learning_rate": 7.80373831775701e-06, | |
| "loss": 0.9887, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.30493426322937, | |
| "learning_rate": 7.336448598130842e-06, | |
| "loss": 0.9634, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 2.5083048343658447, | |
| "learning_rate": 6.869158878504674e-06, | |
| "loss": 0.9366, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.4772727272727273, | |
| "grad_norm": 2.103290557861328, | |
| "learning_rate": 6.401869158878505e-06, | |
| "loss": 0.9368, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5909090909090908, | |
| "grad_norm": 1.9462023973464966, | |
| "learning_rate": 5.9345794392523374e-06, | |
| "loss": 0.9041, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.7045454545454546, | |
| "grad_norm": 2.3276283740997314, | |
| "learning_rate": 5.467289719626169e-06, | |
| "loss": 0.8774, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 2.0134103298187256, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8964, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.9318181818181817, | |
| "grad_norm": 2.1925437450408936, | |
| "learning_rate": 4.532710280373832e-06, | |
| "loss": 0.8861, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.8932154774665833, | |
| "eval_runtime": 528.5967, | |
| "eval_samples_per_second": 5.267, | |
| "eval_steps_per_second": 0.165, | |
| "eval_wer": 0.7331001192128046, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 2.0454545454545454, | |
| "grad_norm": 1.884445309638977, | |
| "learning_rate": 4.065420560747663e-06, | |
| "loss": 0.8624, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.159090909090909, | |
| "grad_norm": 1.7217223644256592, | |
| "learning_rate": 3.5981308411214953e-06, | |
| "loss": 0.8, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.2727272727272725, | |
| "grad_norm": 2.1738855838775635, | |
| "learning_rate": 3.1308411214953272e-06, | |
| "loss": 0.7747, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.3863636363636362, | |
| "grad_norm": 1.8098112344741821, | |
| "learning_rate": 2.6635514018691587e-06, | |
| "loss": 0.8097, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.9226168394088745, | |
| "learning_rate": 2.1962616822429906e-06, | |
| "loss": 0.7781, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.6136363636363638, | |
| "grad_norm": 1.8235868215560913, | |
| "learning_rate": 1.7289719626168225e-06, | |
| "loss": 0.78, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 1.8217195272445679, | |
| "learning_rate": 1.2616822429906543e-06, | |
| "loss": 0.7737, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.840909090909091, | |
| "grad_norm": 1.7603012323379517, | |
| "learning_rate": 7.94392523364486e-07, | |
| "loss": 0.7679, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.9545454545454546, | |
| "grad_norm": 1.8884203433990479, | |
| "learning_rate": 3.2710280373831776e-07, | |
| "loss": 0.779, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.8655255436897278, | |
| "eval_runtime": 520.291, | |
| "eval_samples_per_second": 5.351, | |
| "eval_steps_per_second": 0.167, | |
| "eval_wer": 0.7213644980359202, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 264, | |
| "total_flos": 3.4488281344690356e+19, | |
| "train_loss": 1.3980728189150493, | |
| "train_runtime": 2437.9898, | |
| "train_samples_per_second": 13.714, | |
| "train_steps_per_second": 0.108 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 264, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.4488281344690356e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |