| { | |
| "best_metric": 0.7843137254901961, | |
| "best_model_checkpoint": "vit-base-patch16-224-ve-U10-12\\checkpoint-108", | |
| "epoch": 11.612903225806452, | |
| "eval_steps": 500, | |
| "global_step": 180, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 5.467836257309941e-05, | |
| "loss": 1.3659, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_accuracy": 0.47058823529411764, | |
| "eval_loss": 1.2287312746047974, | |
| "eval_runtime": 0.9667, | |
| "eval_samples_per_second": 52.757, | |
| "eval_steps_per_second": 2.069, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 5.1461988304093566e-05, | |
| "loss": 1.2407, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 4.824561403508772e-05, | |
| "loss": 1.1063, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6862745098039216, | |
| "eval_loss": 1.0233995914459229, | |
| "eval_runtime": 0.8689, | |
| "eval_samples_per_second": 58.697, | |
| "eval_steps_per_second": 2.302, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 4.502923976608187e-05, | |
| "loss": 0.9468, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.6862745098039216, | |
| "eval_loss": 0.8665008544921875, | |
| "eval_runtime": 0.8118, | |
| "eval_samples_per_second": 62.826, | |
| "eval_steps_per_second": 2.464, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 4.1812865497076025e-05, | |
| "loss": 0.8276, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 3.859649122807018e-05, | |
| "loss": 0.6825, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7058823529411765, | |
| "eval_loss": 0.7481548190116882, | |
| "eval_runtime": 0.7724, | |
| "eval_samples_per_second": 66.031, | |
| "eval_steps_per_second": 2.589, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 3.5380116959064324e-05, | |
| "loss": 0.5534, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "eval_accuracy": 0.7058823529411765, | |
| "eval_loss": 0.7608624696731567, | |
| "eval_runtime": 0.8338, | |
| "eval_samples_per_second": 61.167, | |
| "eval_steps_per_second": 2.399, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.4733, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 2.894736842105263e-05, | |
| "loss": 0.4019, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7254901960784313, | |
| "eval_loss": 0.7092276215553284, | |
| "eval_runtime": 0.7947, | |
| "eval_samples_per_second": 64.177, | |
| "eval_steps_per_second": 2.517, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 2.5730994152046783e-05, | |
| "loss": 0.3323, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.7843137254901961, | |
| "eval_loss": 0.662259042263031, | |
| "eval_runtime": 0.7839, | |
| "eval_samples_per_second": 65.057, | |
| "eval_steps_per_second": 2.551, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 2.2514619883040936e-05, | |
| "loss": 0.2759, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 1.929824561403509e-05, | |
| "loss": 0.2743, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7058823529411765, | |
| "eval_loss": 0.7407493591308594, | |
| "eval_runtime": 0.7753, | |
| "eval_samples_per_second": 65.783, | |
| "eval_steps_per_second": 2.58, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 1.608187134502924e-05, | |
| "loss": 0.2411, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_accuracy": 0.7647058823529411, | |
| "eval_loss": 0.6248741745948792, | |
| "eval_runtime": 0.8159, | |
| "eval_samples_per_second": 62.507, | |
| "eval_steps_per_second": 2.451, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 1.2865497076023392e-05, | |
| "loss": 0.2379, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 9.649122807017545e-06, | |
| "loss": 0.2021, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7450980392156863, | |
| "eval_loss": 0.7221617102622986, | |
| "eval_runtime": 0.7603, | |
| "eval_samples_per_second": 67.081, | |
| "eval_steps_per_second": 2.631, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 6.432748538011696e-06, | |
| "loss": 0.195, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 3.216374269005848e-06, | |
| "loss": 0.1925, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.7058823529411765, | |
| "eval_loss": 0.7807687520980835, | |
| "eval_runtime": 0.7671, | |
| "eval_samples_per_second": 66.482, | |
| "eval_steps_per_second": 2.607, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 0.0, | |
| "loss": 0.1748, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "eval_accuracy": 0.7254901960784313, | |
| "eval_loss": 0.7451220750808716, | |
| "eval_runtime": 0.7609, | |
| "eval_samples_per_second": 67.03, | |
| "eval_steps_per_second": 2.629, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "step": 180, | |
| "total_flos": 1.7709561831334625e+18, | |
| "train_loss": 0.5402387758096059, | |
| "train_runtime": 360.9103, | |
| "train_samples_per_second": 65.401, | |
| "train_steps_per_second": 0.499 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 12, | |
| "save_steps": 500, | |
| "total_flos": 1.7709561831334625e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |