| { | |
| "best_metric": 0.05225329101085663, | |
| "best_model_checkpoint": "./vit-base-catsVSdogs-demo-v5/checkpoint-100", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00019, | |
| "loss": 0.4108, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00018, | |
| "loss": 0.1022, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00017, | |
| "loss": 0.2383, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00016, | |
| "loss": 0.1444, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.0812, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0268, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 0.0375, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0088, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.00011000000000000002, | |
| "loss": 0.0227, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0337, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.98, | |
| "eval_loss": 0.05225329101085663, | |
| "eval_runtime": 164.0992, | |
| "eval_samples_per_second": 1.219, | |
| "eval_steps_per_second": 0.152, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0065, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0055, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0179, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 6e-05, | |
| "loss": 0.016, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0314, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0067, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 3e-05, | |
| "loss": 0.004, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0041, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0039, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0038, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.985, | |
| "eval_loss": 0.059128113090991974, | |
| "eval_runtime": 162.5287, | |
| "eval_samples_per_second": 1.231, | |
| "eval_steps_per_second": 0.154, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 200, | |
| "total_flos": 2.479743667666944e+17, | |
| "train_loss": 0.06029550410807133, | |
| "train_runtime": 7671.8377, | |
| "train_samples_per_second": 0.417, | |
| "train_steps_per_second": 0.026 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "total_flos": 2.479743667666944e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |