| { | |
| "best_metric": 0.8807692307692307, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned/checkpoint-108", | |
| "epoch": 2.938775510204082, | |
| "eval_steps": 500, | |
| "global_step": 108, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.272108843537415, | |
| "grad_norm": 8.761322975158691, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 1.8625, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.54421768707483, | |
| "grad_norm": 13.928447723388672, | |
| "learning_rate": 4.536082474226804e-05, | |
| "loss": 1.5024, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8163265306122449, | |
| "grad_norm": 14.840970993041992, | |
| "learning_rate": 4.020618556701031e-05, | |
| "loss": 1.0835, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9795918367346939, | |
| "eval_accuracy": 0.6884615384615385, | |
| "eval_loss": 0.7885341048240662, | |
| "eval_runtime": 4.1617, | |
| "eval_samples_per_second": 124.947, | |
| "eval_steps_per_second": 4.085, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.08843537414966, | |
| "grad_norm": 13.623029708862305, | |
| "learning_rate": 3.5051546391752576e-05, | |
| "loss": 0.857, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3605442176870748, | |
| "grad_norm": 18.793283462524414, | |
| "learning_rate": 2.9896907216494846e-05, | |
| "loss": 0.7227, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.6326530612244898, | |
| "grad_norm": 9.9773530960083, | |
| "learning_rate": 2.4742268041237116e-05, | |
| "loss": 0.6269, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 10.044248580932617, | |
| "learning_rate": 1.9587628865979382e-05, | |
| "loss": 0.6076, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.9863945578231292, | |
| "eval_accuracy": 0.8096153846153846, | |
| "eval_loss": 0.5072689652442932, | |
| "eval_runtime": 4.141, | |
| "eval_samples_per_second": 125.573, | |
| "eval_steps_per_second": 4.105, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.17687074829932, | |
| "grad_norm": 11.315401077270508, | |
| "learning_rate": 1.4432989690721649e-05, | |
| "loss": 0.5397, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.4489795918367347, | |
| "grad_norm": 7.718348503112793, | |
| "learning_rate": 9.278350515463918e-06, | |
| "loss": 0.5192, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.7210884353741496, | |
| "grad_norm": 5.907692909240723, | |
| "learning_rate": 4.123711340206186e-06, | |
| "loss": 0.499, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.938775510204082, | |
| "eval_accuracy": 0.8807692307692307, | |
| "eval_loss": 0.3335144817829132, | |
| "eval_runtime": 4.0967, | |
| "eval_samples_per_second": 126.933, | |
| "eval_steps_per_second": 4.15, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.938775510204082, | |
| "step": 108, | |
| "total_flos": 3.424639439220572e+17, | |
| "train_loss": 0.8488516895859329, | |
| "train_runtime": 194.721, | |
| "train_samples_per_second": 72.103, | |
| "train_steps_per_second": 0.555 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 108, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 3.424639439220572e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |