| { | |
| "best_metric": 0.7530120481927711, | |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-finetuned-gardner-te-max/checkpoint-129", | |
| "epoch": 18.72340425531915, | |
| "eval_steps": 500, | |
| "global_step": 220, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 1.0943, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_accuracy": 0.6325301204819277, | |
| "eval_loss": 1.0749578475952148, | |
| "eval_runtime": 8.5945, | |
| "eval_samples_per_second": 19.315, | |
| "eval_steps_per_second": 0.698, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.9996, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_accuracy": 0.6325301204819277, | |
| "eval_loss": 0.8011154532432556, | |
| "eval_runtime": 2.9146, | |
| "eval_samples_per_second": 56.955, | |
| "eval_steps_per_second": 2.059, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.797979797979798e-05, | |
| "loss": 0.7731, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.6325301204819277, | |
| "eval_loss": 0.7181705236434937, | |
| "eval_runtime": 2.8916, | |
| "eval_samples_per_second": 57.408, | |
| "eval_steps_per_second": 2.075, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.7564, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6325301204819277, | |
| "eval_loss": 0.7109255790710449, | |
| "eval_runtime": 2.8917, | |
| "eval_samples_per_second": 57.405, | |
| "eval_steps_per_second": 2.075, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 4.292929292929293e-05, | |
| "loss": 0.7331, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_accuracy": 0.6325301204819277, | |
| "eval_loss": 0.7025781869888306, | |
| "eval_runtime": 2.9658, | |
| "eval_samples_per_second": 55.972, | |
| "eval_steps_per_second": 2.023, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 4.0404040404040405e-05, | |
| "loss": 0.743, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 3.787878787878788e-05, | |
| "loss": 0.7336, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "eval_accuracy": 0.6325301204819277, | |
| "eval_loss": 0.6847715973854065, | |
| "eval_runtime": 2.9186, | |
| "eval_samples_per_second": 56.877, | |
| "eval_steps_per_second": 2.056, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 3.535353535353535e-05, | |
| "loss": 0.7305, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_accuracy": 0.6325301204819277, | |
| "eval_loss": 0.6938183307647705, | |
| "eval_runtime": 2.945, | |
| "eval_samples_per_second": 56.367, | |
| "eval_steps_per_second": 2.037, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 3.282828282828283e-05, | |
| "loss": 0.7314, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6325301204819277, | |
| "eval_loss": 0.654893696308136, | |
| "eval_runtime": 2.8829, | |
| "eval_samples_per_second": 57.58, | |
| "eval_steps_per_second": 2.081, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 3.0303030303030306e-05, | |
| "loss": 0.6905, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "eval_accuracy": 0.6867469879518072, | |
| "eval_loss": 0.6364297270774841, | |
| "eval_runtime": 2.9426, | |
| "eval_samples_per_second": 56.413, | |
| "eval_steps_per_second": 2.039, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.7315, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 9.96, | |
| "eval_accuracy": 0.6686746987951807, | |
| "eval_loss": 0.6223430037498474, | |
| "eval_runtime": 2.8889, | |
| "eval_samples_per_second": 57.46, | |
| "eval_steps_per_second": 2.077, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 10.21, | |
| "learning_rate": 2.5252525252525256e-05, | |
| "loss": 0.6839, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "eval_accuracy": 0.7530120481927711, | |
| "eval_loss": 0.6527661085128784, | |
| "eval_runtime": 2.9492, | |
| "eval_samples_per_second": 56.286, | |
| "eval_steps_per_second": 2.034, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.7044, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 11.91, | |
| "learning_rate": 2.0202020202020203e-05, | |
| "loss": 0.6931, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7409638554216867, | |
| "eval_loss": 0.6209445595741272, | |
| "eval_runtime": 2.9113, | |
| "eval_samples_per_second": 57.019, | |
| "eval_steps_per_second": 2.061, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 12.77, | |
| "learning_rate": 1.7676767676767676e-05, | |
| "loss": 0.6705, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 12.94, | |
| "eval_accuracy": 0.7168674698795181, | |
| "eval_loss": 0.6295837759971619, | |
| "eval_runtime": 2.9661, | |
| "eval_samples_per_second": 55.965, | |
| "eval_steps_per_second": 2.023, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 13.62, | |
| "learning_rate": 1.5151515151515153e-05, | |
| "loss": 0.7227, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 13.96, | |
| "eval_accuracy": 0.7108433734939759, | |
| "eval_loss": 0.6039193272590637, | |
| "eval_runtime": 2.9739, | |
| "eval_samples_per_second": 55.819, | |
| "eval_steps_per_second": 2.018, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 1.2626262626262628e-05, | |
| "loss": 0.6695, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "eval_accuracy": 0.7530120481927711, | |
| "eval_loss": 0.6048843264579773, | |
| "eval_runtime": 2.9528, | |
| "eval_samples_per_second": 56.217, | |
| "eval_steps_per_second": 2.032, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 15.32, | |
| "learning_rate": 1.0101010101010101e-05, | |
| "loss": 0.6981, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7048192771084337, | |
| "eval_loss": 0.5965093374252319, | |
| "eval_runtime": 2.9154, | |
| "eval_samples_per_second": 56.938, | |
| "eval_steps_per_second": 2.058, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 16.17, | |
| "learning_rate": 7.5757575757575764e-06, | |
| "loss": 0.6566, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "eval_accuracy": 0.7409638554216867, | |
| "eval_loss": 0.6111359000205994, | |
| "eval_runtime": 2.8973, | |
| "eval_samples_per_second": 57.294, | |
| "eval_steps_per_second": 2.071, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "learning_rate": 5.050505050505051e-06, | |
| "loss": 0.6908, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 17.87, | |
| "learning_rate": 2.5252525252525253e-06, | |
| "loss": 0.6828, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "eval_accuracy": 0.7530120481927711, | |
| "eval_loss": 0.5969204902648926, | |
| "eval_runtime": 2.8702, | |
| "eval_samples_per_second": 57.835, | |
| "eval_steps_per_second": 2.09, | |
| "step": 211 | |
| }, | |
| { | |
| "epoch": 18.72, | |
| "learning_rate": 0.0, | |
| "loss": 0.6632, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 18.72, | |
| "eval_accuracy": 0.7530120481927711, | |
| "eval_loss": 0.5946705937385559, | |
| "eval_runtime": 2.8876, | |
| "eval_samples_per_second": 57.486, | |
| "eval_steps_per_second": 2.078, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 18.72, | |
| "step": 220, | |
| "total_flos": 9.056508197685166e+17, | |
| "train_loss": 0.7387422084808349, | |
| "train_runtime": 1241.7001, | |
| "train_samples_per_second": 23.935, | |
| "train_steps_per_second": 0.177 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 220, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 9.056508197685166e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |