| { | |
| "best_metric": 0.6416184971098265, | |
| "best_model_checkpoint": "resnet-50-finetuned-eurosat/checkpoint-97", | |
| "epoch": 2.9846153846153847, | |
| "eval_steps": 500, | |
| "global_step": 291, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.10256410256410256, | |
| "grad_norm": 0.4352456331253052, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.6926, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.20512820512820512, | |
| "grad_norm": 0.80888432264328, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6876, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 0.7256971597671509, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6791, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.41025641025641024, | |
| "grad_norm": 0.7818945050239563, | |
| "learning_rate": 4.8084291187739464e-05, | |
| "loss": 0.6602, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5128205128205128, | |
| "grad_norm": 0.5908713340759277, | |
| "learning_rate": 4.616858237547893e-05, | |
| "loss": 0.6622, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 0.570097804069519, | |
| "learning_rate": 4.4252873563218394e-05, | |
| "loss": 0.6582, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.717948717948718, | |
| "grad_norm": 0.7440472841262817, | |
| "learning_rate": 4.2337164750957856e-05, | |
| "loss": 0.646, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8205128205128205, | |
| "grad_norm": 0.5741876363754272, | |
| "learning_rate": 4.0421455938697324e-05, | |
| "loss": 0.6599, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 0.6604257822036743, | |
| "learning_rate": 3.850574712643678e-05, | |
| "loss": 0.6486, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.9948717948717949, | |
| "eval_accuracy": 0.6416184971098265, | |
| "eval_loss": 0.6518892645835876, | |
| "eval_runtime": 17.6276, | |
| "eval_samples_per_second": 78.513, | |
| "eval_steps_per_second": 2.496, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 1.0256410256410255, | |
| "grad_norm": 0.8966683149337769, | |
| "learning_rate": 3.659003831417625e-05, | |
| "loss": 0.6356, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1282051282051282, | |
| "grad_norm": 0.7184256911277771, | |
| "learning_rate": 3.467432950191571e-05, | |
| "loss": 0.6409, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 0.9605466723442078, | |
| "learning_rate": 3.275862068965517e-05, | |
| "loss": 0.6452, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.6847850680351257, | |
| "learning_rate": 3.084291187739464e-05, | |
| "loss": 0.6562, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.435897435897436, | |
| "grad_norm": 0.7334717512130737, | |
| "learning_rate": 2.89272030651341e-05, | |
| "loss": 0.631, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.962297797203064, | |
| "learning_rate": 2.7011494252873566e-05, | |
| "loss": 0.6496, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.641025641025641, | |
| "grad_norm": 0.8398529887199402, | |
| "learning_rate": 2.5095785440613027e-05, | |
| "loss": 0.6555, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7435897435897436, | |
| "grad_norm": 0.8672203421592712, | |
| "learning_rate": 2.3180076628352492e-05, | |
| "loss": 0.6496, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 0.9078596830368042, | |
| "learning_rate": 2.1264367816091954e-05, | |
| "loss": 0.6404, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9487179487179487, | |
| "grad_norm": 1.1112207174301147, | |
| "learning_rate": 1.934865900383142e-05, | |
| "loss": 0.6518, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6416184971098265, | |
| "eval_loss": 0.6501337885856628, | |
| "eval_runtime": 15.5449, | |
| "eval_samples_per_second": 89.032, | |
| "eval_steps_per_second": 2.831, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.051282051282051, | |
| "grad_norm": 1.176990032196045, | |
| "learning_rate": 1.743295019157088e-05, | |
| "loss": 0.6564, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 0.9637936353683472, | |
| "learning_rate": 1.5517241379310346e-05, | |
| "loss": 0.6496, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.2564102564102564, | |
| "grad_norm": 0.6943733096122742, | |
| "learning_rate": 1.360153256704981e-05, | |
| "loss": 0.6496, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.358974358974359, | |
| "grad_norm": 0.8282309770584106, | |
| "learning_rate": 1.1685823754789272e-05, | |
| "loss": 0.6412, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 0.7222530841827393, | |
| "learning_rate": 9.770114942528738e-06, | |
| "loss": 0.6454, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.564102564102564, | |
| "grad_norm": 0.9417911171913147, | |
| "learning_rate": 7.854406130268199e-06, | |
| "loss": 0.6406, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.8182082176208496, | |
| "learning_rate": 5.938697318007663e-06, | |
| "loss": 0.6594, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 0.8655200600624084, | |
| "learning_rate": 4.022988505747127e-06, | |
| "loss": 0.6452, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.871794871794872, | |
| "grad_norm": 0.6346496939659119, | |
| "learning_rate": 2.1072796934865904e-06, | |
| "loss": 0.6435, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.9743589743589745, | |
| "grad_norm": 0.5316995978355408, | |
| "learning_rate": 1.9157088122605365e-07, | |
| "loss": 0.6562, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.9846153846153847, | |
| "eval_accuracy": 0.6416184971098265, | |
| "eval_loss": 0.6498913168907166, | |
| "eval_runtime": 16.5397, | |
| "eval_samples_per_second": 83.678, | |
| "eval_steps_per_second": 2.66, | |
| "step": 291 | |
| }, | |
| { | |
| "epoch": 2.9846153846153847, | |
| "step": 291, | |
| "total_flos": 7.898720659019366e+17, | |
| "train_loss": 0.6530564023866686, | |
| "train_runtime": 529.8812, | |
| "train_samples_per_second": 70.51, | |
| "train_steps_per_second": 0.549 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 291, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 7.898720659019366e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |