| { | |
| "best_metric": 0.7334273624823695, | |
| "best_model_checkpoint": "vit-base-patch16-224-in21k-harm-C/checkpoint-45", | |
| "epoch": 7.822222222222222, | |
| "global_step": 88, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00011848101265822785, | |
| "loss": 0.9489, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_accuracy": 0.6629055007052186, | |
| "eval_loss": 0.7731553316116333, | |
| "eval_runtime": 11.023, | |
| "eval_samples_per_second": 64.32, | |
| "eval_steps_per_second": 1.089, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00010329113924050633, | |
| "loss": 0.7902, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_accuracy": 0.7009873060648801, | |
| "eval_loss": 0.7204822897911072, | |
| "eval_runtime": 10.8566, | |
| "eval_samples_per_second": 65.306, | |
| "eval_steps_per_second": 1.105, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 8.810126582278482e-05, | |
| "loss": 0.7452, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_accuracy": 0.7207334273624824, | |
| "eval_loss": 0.6992235779762268, | |
| "eval_runtime": 10.7875, | |
| "eval_samples_per_second": 65.724, | |
| "eval_steps_per_second": 1.112, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 7.29113924050633e-05, | |
| "loss": 0.6899, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7334273624823695, | |
| "eval_loss": 0.6885057687759399, | |
| "eval_runtime": 11.0176, | |
| "eval_samples_per_second": 64.351, | |
| "eval_steps_per_second": 1.089, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 5.7721518987341775e-05, | |
| "loss": 0.6788, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "eval_accuracy": 0.7136812411847673, | |
| "eval_loss": 0.6880338788032532, | |
| "eval_runtime": 10.9845, | |
| "eval_samples_per_second": 64.545, | |
| "eval_steps_per_second": 1.092, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 4.253164556962025e-05, | |
| "loss": 0.6116, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "eval_accuracy": 0.7122708039492243, | |
| "eval_loss": 0.6894987225532532, | |
| "eval_runtime": 10.9278, | |
| "eval_samples_per_second": 64.88, | |
| "eval_steps_per_second": 1.098, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 2.7341772151898733e-05, | |
| "loss": 0.5875, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "eval_accuracy": 0.7193229901269393, | |
| "eval_loss": 0.6877116560935974, | |
| "eval_runtime": 10.9862, | |
| "eval_samples_per_second": 64.535, | |
| "eval_steps_per_second": 1.092, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 1.2151898734177216e-05, | |
| "loss": 0.562, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.82, | |
| "eval_accuracy": 0.7136812411847673, | |
| "eval_loss": 0.6862055659294128, | |
| "eval_runtime": 10.9399, | |
| "eval_samples_per_second": 64.809, | |
| "eval_steps_per_second": 1.097, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 7.82, | |
| "step": 88, | |
| "total_flos": 1.721344993041365e+18, | |
| "train_loss": 0.6865291487086903, | |
| "train_runtime": 716.0754, | |
| "train_samples_per_second": 31.673, | |
| "train_steps_per_second": 0.123 | |
| } | |
| ], | |
| "max_steps": 88, | |
| "num_train_epochs": 8, | |
| "total_flos": 1.721344993041365e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |