| { | |
| "best_metric": 27.7224217287947, | |
| "best_model_checkpoint": "codefactory4791/whisper-small-medicalv2/checkpoint-4500", | |
| "epoch": 20.408163265306122, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.92e-06, | |
| "loss": 1.5644, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.5293, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_loss": 0.48172396421432495, | |
| "eval_runtime": 325.9765, | |
| "eval_samples_per_second": 4.23, | |
| "eval_steps_per_second": 0.135, | |
| "eval_wer": 46.628900915482276, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 9.742105263157897e-06, | |
| "loss": 0.5005, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 9.478947368421053e-06, | |
| "loss": 0.4125, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 0.5011524558067322, | |
| "eval_runtime": 303.4033, | |
| "eval_samples_per_second": 4.545, | |
| "eval_steps_per_second": 0.145, | |
| "eval_wer": 32.163269435031104, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 9.215789473684211e-06, | |
| "loss": 0.3928, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 8.95263157894737e-06, | |
| "loss": 0.3263, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "eval_loss": 0.5298507809638977, | |
| "eval_runtime": 296.4623, | |
| "eval_samples_per_second": 4.652, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 29.558444185928884, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 8.689473684210526e-06, | |
| "loss": 0.2629, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 8.426315789473684e-06, | |
| "loss": 0.2399, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_loss": 0.5742943286895752, | |
| "eval_runtime": 290.0311, | |
| "eval_samples_per_second": 4.755, | |
| "eval_steps_per_second": 0.152, | |
| "eval_wer": 28.157402255829243, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 8.163157894736842e-06, | |
| "loss": 0.1817, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 7.9e-06, | |
| "loss": 0.1563, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "eval_loss": 0.6200889348983765, | |
| "eval_runtime": 297.5485, | |
| "eval_samples_per_second": 4.635, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 39.52253300288301, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 7.636842105263159e-06, | |
| "loss": 0.1411, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 7.373684210526316e-06, | |
| "loss": 0.095, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "eval_loss": 0.6449038982391357, | |
| "eval_runtime": 288.5774, | |
| "eval_samples_per_second": 4.779, | |
| "eval_steps_per_second": 0.152, | |
| "eval_wer": 28.481108694552624, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 7.110526315789474e-06, | |
| "loss": 0.0869, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 6.8473684210526325e-06, | |
| "loss": 0.0767, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "eval_loss": 0.671624481678009, | |
| "eval_runtime": 290.9537, | |
| "eval_samples_per_second": 4.74, | |
| "eval_steps_per_second": 0.151, | |
| "eval_wer": 29.841687319811843, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 6.58421052631579e-06, | |
| "loss": 0.0598, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 10.2, | |
| "learning_rate": 6.321052631578948e-06, | |
| "loss": 0.0537, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.2, | |
| "eval_loss": 0.71231609582901, | |
| "eval_runtime": 295.3003, | |
| "eval_samples_per_second": 4.67, | |
| "eval_steps_per_second": 0.149, | |
| "eval_wer": 28.354660866926306, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.84, | |
| "learning_rate": 6.057894736842106e-06, | |
| "loss": 0.0486, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 11.48, | |
| "learning_rate": 5.794736842105264e-06, | |
| "loss": 0.0406, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.48, | |
| "eval_loss": 0.72938472032547, | |
| "eval_runtime": 288.8463, | |
| "eval_samples_per_second": 4.774, | |
| "eval_steps_per_second": 0.152, | |
| "eval_wer": 27.7224217287947, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.12, | |
| "learning_rate": 5.531578947368421e-06, | |
| "loss": 0.0376, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "learning_rate": 5.268421052631579e-06, | |
| "loss": 0.0314, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "eval_loss": 0.7565638422966003, | |
| "eval_runtime": 294.0019, | |
| "eval_samples_per_second": 4.69, | |
| "eval_steps_per_second": 0.15, | |
| "eval_wer": 27.950027818522077, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.39, | |
| "learning_rate": 5.005263157894737e-06, | |
| "loss": 0.0278, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 4.7421052631578954e-06, | |
| "loss": 0.026, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "eval_loss": 0.7726335525512695, | |
| "eval_runtime": 292.0072, | |
| "eval_samples_per_second": 4.722, | |
| "eval_steps_per_second": 0.151, | |
| "eval_wer": 28.106823124778717, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.67, | |
| "learning_rate": 4.478947368421054e-06, | |
| "loss": 0.0222, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "learning_rate": 4.215789473684211e-06, | |
| "loss": 0.0212, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "eval_loss": 0.757480263710022, | |
| "eval_runtime": 302.5701, | |
| "eval_samples_per_second": 4.558, | |
| "eval_steps_per_second": 0.145, | |
| "eval_wer": 31.252845076121595, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.94, | |
| "learning_rate": 3.952631578947368e-06, | |
| "loss": 0.0191, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 16.58, | |
| "learning_rate": 3.6894736842105265e-06, | |
| "loss": 0.016, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 16.58, | |
| "eval_loss": 0.7838702201843262, | |
| "eval_runtime": 293.8696, | |
| "eval_samples_per_second": 4.693, | |
| "eval_steps_per_second": 0.15, | |
| "eval_wer": 28.34960295382125, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.22, | |
| "learning_rate": 3.4263157894736842e-06, | |
| "loss": 0.0154, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 3.1631578947368424e-06, | |
| "loss": 0.0132, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "eval_loss": 0.7901157140731812, | |
| "eval_runtime": 300.427, | |
| "eval_samples_per_second": 4.59, | |
| "eval_steps_per_second": 0.146, | |
| "eval_wer": 28.98184209195286, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.49, | |
| "learning_rate": 2.9e-06, | |
| "loss": 0.0114, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "learning_rate": 2.6368421052631584e-06, | |
| "loss": 0.0112, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "eval_loss": 0.8057935833930969, | |
| "eval_runtime": 299.4809, | |
| "eval_samples_per_second": 4.605, | |
| "eval_steps_per_second": 0.147, | |
| "eval_wer": 29.027363309898334, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.77, | |
| "learning_rate": 2.373684210526316e-06, | |
| "loss": 0.0089, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 20.41, | |
| "learning_rate": 2.110526315789474e-06, | |
| "loss": 0.0078, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 20.41, | |
| "eval_loss": 0.8125391006469727, | |
| "eval_runtime": 297.7542, | |
| "eval_samples_per_second": 4.631, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 29.593849577664255, | |
| "step": 8000 | |
| } | |
| ], | |
| "max_steps": 10000, | |
| "num_train_epochs": 26, | |
| "total_flos": 1.47495998103552e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |