| { | |
| "best_metric": 9.103012781497261, | |
| "best_model_checkpoint": "./Whisper-squeezeformer-v4\\checkpoint-33000", | |
| "epoch": 14.4, | |
| "eval_steps": 3000, | |
| "global_step": 36000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 14.352978706359863, | |
| "learning_rate": 9.976666666666667e-06, | |
| "loss": 3.89, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.287811756134033, | |
| "eval_runtime": 2286.1099, | |
| "eval_samples_per_second": 2.187, | |
| "eval_steps_per_second": 0.273, | |
| "eval_wer": 114.50152344831994, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 9.717866897583008, | |
| "learning_rate": 8.004666666666668e-06, | |
| "loss": 1.1579, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7946763634681702, | |
| "eval_runtime": 2115.529, | |
| "eval_samples_per_second": 2.363, | |
| "eval_steps_per_second": 0.295, | |
| "eval_wer": 42.057827115249395, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 7.917266845703125, | |
| "learning_rate": 6.004666666666668e-06, | |
| "loss": 0.3888, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.7378555536270142, | |
| "eval_runtime": 2241.804, | |
| "eval_samples_per_second": 2.23, | |
| "eval_steps_per_second": 0.279, | |
| "eval_wer": 36.931370251209174, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 5.832924842834473, | |
| "learning_rate": 4.004666666666667e-06, | |
| "loss": 0.2242, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.7416579127311707, | |
| "eval_runtime": 2509.1854, | |
| "eval_samples_per_second": 1.993, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 35.91715850254618, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 10.248781204223633, | |
| "learning_rate": 5.558888888888889e-06, | |
| "loss": 0.5221, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.6810700297355652, | |
| "eval_runtime": 2258.0308, | |
| "eval_samples_per_second": 2.214, | |
| "eval_steps_per_second": 0.277, | |
| "eval_wer": 32.78077259071443, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 7.38523530960083, | |
| "learning_rate": 4.448148148148149e-06, | |
| "loss": 0.324, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.671556830406189, | |
| "eval_runtime": 2259.7666, | |
| "eval_samples_per_second": 2.213, | |
| "eval_steps_per_second": 0.277, | |
| "eval_wer": 32.045682142628856, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 5.77803373336792, | |
| "learning_rate": 3.337777777777778e-06, | |
| "loss": 0.2034, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.684516429901123, | |
| "eval_runtime": 2285.8284, | |
| "eval_samples_per_second": 2.187, | |
| "eval_steps_per_second": 0.273, | |
| "eval_wer": 32.007329597511344, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "grad_norm": 3.751537322998047, | |
| "learning_rate": 2.2274074074074075e-06, | |
| "loss": 0.2177, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "eval_loss": 0.19905297458171844, | |
| "eval_runtime": 1258.2631, | |
| "eval_samples_per_second": 2.082, | |
| "eval_steps_per_second": 0.261, | |
| "eval_wer": 10.862370663420572, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 10.8, | |
| "grad_norm": 2.6073410511016846, | |
| "learning_rate": 1.1166666666666666e-06, | |
| "loss": 0.127, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 10.8, | |
| "eval_loss": 0.1856304407119751, | |
| "eval_runtime": 1152.5547, | |
| "eval_samples_per_second": 2.273, | |
| "eval_steps_per_second": 0.285, | |
| "eval_wer": 10.548539257455873, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 2.8211987018585205, | |
| "learning_rate": 5.555555555555556e-09, | |
| "loss": 0.0909, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.18379394710063934, | |
| "eval_runtime": 1196.9273, | |
| "eval_samples_per_second": 2.189, | |
| "eval_steps_per_second": 0.274, | |
| "eval_wer": 9.591828971393792, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 13.2, | |
| "grad_norm": 1.6552714109420776, | |
| "learning_rate": 9.13939393939394e-07, | |
| "loss": 0.0785, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 13.2, | |
| "eval_loss": 0.18493999540805817, | |
| "eval_runtime": 1318.1594, | |
| "eval_samples_per_second": 1.988, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 9.103012781497261, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "grad_norm": 2.0688605308532715, | |
| "learning_rate": 5.151515151515151e-09, | |
| "loss": 0.0595, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "eval_loss": 0.1860339492559433, | |
| "eval_runtime": 1345.7303, | |
| "eval_samples_per_second": 1.947, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 9.129640900791236, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "step": 36000, | |
| "total_flos": 2.495852642304e+20, | |
| "train_loss": 0.011502729203965929, | |
| "train_runtime": 50610.9609, | |
| "train_samples_per_second": 14.226, | |
| "train_steps_per_second": 0.711 | |
| } | |
| ], | |
| "logging_steps": 3000, | |
| "max_steps": 36000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 3000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.495852642304e+20, | |
| "train_batch_size": 20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |