| { | |
| "best_metric": 0.9608938547486033, | |
| "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-CT\\checkpoint-87", | |
| "epoch": 2.9491525423728815, | |
| "eval_steps": 500, | |
| "global_step": 87, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3389830508474576, | |
| "grad_norm": 1.1623811721801758, | |
| "learning_rate": 4.935897435897436e-05, | |
| "loss": 1.3079, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6779661016949152, | |
| "grad_norm": 1.0423816442489624, | |
| "learning_rate": 4.294871794871795e-05, | |
| "loss": 0.9129, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9830508474576272, | |
| "eval_accuracy": 0.9329608938547486, | |
| "eval_loss": 0.45367431640625, | |
| "eval_runtime": 189.2542, | |
| "eval_samples_per_second": 2.837, | |
| "eval_steps_per_second": 0.09, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.0169491525423728, | |
| "grad_norm": 1.0528792142868042, | |
| "learning_rate": 3.653846153846154e-05, | |
| "loss": 0.5782, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.3559322033898304, | |
| "grad_norm": 1.0269708633422852, | |
| "learning_rate": 3.012820512820513e-05, | |
| "loss": 0.3838, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.694915254237288, | |
| "grad_norm": 1.0555919408798218, | |
| "learning_rate": 2.3717948717948718e-05, | |
| "loss": 0.3054, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9553072625698324, | |
| "eval_loss": 0.24561791121959686, | |
| "eval_runtime": 176.6607, | |
| "eval_samples_per_second": 3.04, | |
| "eval_steps_per_second": 0.096, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.0338983050847457, | |
| "grad_norm": 0.6921894550323486, | |
| "learning_rate": 1.730769230769231e-05, | |
| "loss": 0.2698, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3728813559322033, | |
| "grad_norm": 1.0234838724136353, | |
| "learning_rate": 1.0897435897435898e-05, | |
| "loss": 0.2378, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.711864406779661, | |
| "grad_norm": 1.1442369222640991, | |
| "learning_rate": 4.487179487179488e-06, | |
| "loss": 0.2369, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.9491525423728815, | |
| "eval_accuracy": 0.9608938547486033, | |
| "eval_loss": 0.2215021550655365, | |
| "eval_runtime": 177.0384, | |
| "eval_samples_per_second": 3.033, | |
| "eval_steps_per_second": 0.096, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.9491525423728815, | |
| "step": 87, | |
| "total_flos": 8.603314901828076e+17, | |
| "train_loss": 0.5041907244715197, | |
| "train_runtime": 12510.0722, | |
| "train_samples_per_second": 0.901, | |
| "train_steps_per_second": 0.007 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 87, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 8.603314901828076e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |