| { | |
| "best_metric": 20.920285768001502, | |
| "best_model_checkpoint": "./whisper-base-google-fleurs-pt-br/checkpoint-200", | |
| "epoch": 2.5125628140703515, | |
| "eval_steps": 100, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 5.893749999999999e-06, | |
| "loss": 2.004, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.2299999999999999e-05, | |
| "loss": 1.1337, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.870625e-05, | |
| "loss": 0.8017, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.99875e-05, | |
| "loss": 0.6738, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.3942798376083374, | |
| "eval_runtime": 199.2208, | |
| "eval_samples_per_second": 4.613, | |
| "eval_steps_per_second": 0.291, | |
| "eval_wer": 21.73340853543899, | |
| "eval_wer_normalized": 17.94871794871795, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.9275694444444445e-05, | |
| "loss": 0.5827, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.8563888888888888e-05, | |
| "loss": 0.5512, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.7852083333333335e-05, | |
| "loss": 0.5489, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 1.7140277777777778e-05, | |
| "loss": 0.4816, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 0.3761998414993286, | |
| "eval_runtime": 198.2156, | |
| "eval_samples_per_second": 4.636, | |
| "eval_steps_per_second": 0.293, | |
| "eval_wer": 20.920285768001502, | |
| "eval_wer_normalized": 17.13524570667428, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 1.6428472222222225e-05, | |
| "loss": 0.3379, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.5716666666666668e-05, | |
| "loss": 0.3326, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.500486111111111e-05, | |
| "loss": 0.2885, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 1.4293055555555555e-05, | |
| "loss": 0.2652, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_loss": 0.3871774673461914, | |
| "eval_runtime": 196.5558, | |
| "eval_samples_per_second": 4.676, | |
| "eval_steps_per_second": 0.295, | |
| "eval_wer": 21.188193269411542, | |
| "eval_wer_normalized": 17.282717282717282, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.358125e-05, | |
| "loss": 0.2971, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 1.2869444444444445e-05, | |
| "loss": 0.2761, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 1.215763888888889e-05, | |
| "loss": 0.2682, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 1.1445833333333333e-05, | |
| "loss": 0.2901, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 0.3911706209182739, | |
| "eval_runtime": 200.4943, | |
| "eval_samples_per_second": 4.584, | |
| "eval_steps_per_second": 0.289, | |
| "eval_wer": 21.460800902425266, | |
| "eval_wer_normalized": 17.706103420389134, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 1.0734027777777778e-05, | |
| "loss": 0.1471, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 1.0022222222222222e-05, | |
| "loss": 0.1668, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 9.310416666666667e-06, | |
| "loss": 0.1635, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 8.598611111111112e-06, | |
| "loss": 0.1408, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_loss": 0.4062696099281311, | |
| "eval_runtime": 205.1449, | |
| "eval_samples_per_second": 4.48, | |
| "eval_steps_per_second": 0.283, | |
| "eval_wer": 21.611205113743186, | |
| "eval_wer_normalized": 18.001046572475143, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "step": 500, | |
| "total_flos": 1.03412777877504e+18, | |
| "train_loss": 0.48756390714645387, | |
| "train_runtime": 1663.348, | |
| "train_samples_per_second": 15.391, | |
| "train_steps_per_second": 0.481 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 100, | |
| "total_flos": 1.03412777877504e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |