| { |
| "best_metric": 201.0846405029297, |
| "best_model_checkpoint": "./output/model_output/checkpoint-2000", |
| "epoch": 0.06526882597699274, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.02, |
| "learning_rate": 9.820000000000001e-06, |
| "loss": 340.5683, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.02, |
| "eval_CER": 5.084942084942085, |
| "eval_loss": 333.91064453125, |
| "eval_runtime": 0.3435, |
| "eval_samples_per_second": 29.111, |
| "eval_steps_per_second": 8.733, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.03, |
| "learning_rate": 9.999936439679716e-06, |
| "loss": 306.7755, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.03, |
| "eval_CER": 0.7760617760617761, |
| "eval_loss": 294.28277587890625, |
| "eval_runtime": 0.3348, |
| "eval_samples_per_second": 29.865, |
| "eval_steps_per_second": 8.959, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.05, |
| "learning_rate": 9.999741601099537e-06, |
| "loss": 291.7735, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.05, |
| "eval_CER": 1.0, |
| "eval_loss": 248.63803100585938, |
| "eval_runtime": 0.3203, |
| "eval_samples_per_second": 31.222, |
| "eval_steps_per_second": 9.367, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.07, |
| "learning_rate": 9.999414687126431e-06, |
| "loss": 237.9951, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.07, |
| "eval_CER": 1.0, |
| "eval_loss": 201.0846405029297, |
| "eval_runtime": 0.3248, |
| "eval_samples_per_second": 30.789, |
| "eval_steps_per_second": 9.237, |
| "step": 2000 |
| } |
| ], |
| "max_steps": 306420, |
| "num_train_epochs": 10, |
| "total_flos": 1.648747520322154e+18, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
