| { | |
| "best_metric": 0.7808219178082192, | |
| "best_model_checkpoint": "deit-small-patch16-224-finetuned-piid/checkpoint-307", | |
| "epoch": 19.51219512195122, | |
| "eval_steps": 500, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.4297, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.1537, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_accuracy": 0.547945205479452, | |
| "eval_loss": 1.0005042552947998, | |
| "eval_runtime": 1.0515, | |
| "eval_samples_per_second": 208.282, | |
| "eval_steps_per_second": 26.63, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.8255, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7025, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.593607305936073, | |
| "eval_loss": 0.8481215834617615, | |
| "eval_runtime": 1.2994, | |
| "eval_samples_per_second": 168.538, | |
| "eval_steps_per_second": 21.548, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.6057, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.6581, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.7214611872146118, | |
| "eval_loss": 0.6351255774497986, | |
| "eval_runtime": 1.0449, | |
| "eval_samples_per_second": 209.599, | |
| "eval_steps_per_second": 26.798, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 0.5469, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.5019, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7214611872146118, | |
| "eval_loss": 0.6695594191551208, | |
| "eval_runtime": 1.0755, | |
| "eval_samples_per_second": 203.631, | |
| "eval_steps_per_second": 26.035, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 4.305555555555556e-05, | |
| "loss": 0.5202, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.4708, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "eval_accuracy": 0.7534246575342466, | |
| "eval_loss": 0.5861416459083557, | |
| "eval_runtime": 1.2022, | |
| "eval_samples_per_second": 182.173, | |
| "eval_steps_per_second": 23.292, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "learning_rate": 4.027777777777778e-05, | |
| "loss": 0.5432, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.3647, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.776255707762557, | |
| "eval_loss": 0.5583679676055908, | |
| "eval_runtime": 1.0741, | |
| "eval_samples_per_second": 203.885, | |
| "eval_steps_per_second": 26.068, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3456, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.2973, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_accuracy": 0.7671232876712328, | |
| "eval_loss": 0.5783523321151733, | |
| "eval_runtime": 1.4218, | |
| "eval_samples_per_second": 154.026, | |
| "eval_steps_per_second": 19.693, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.3172, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.2827, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7671232876712328, | |
| "eval_loss": 0.5850614309310913, | |
| "eval_runtime": 1.1505, | |
| "eval_samples_per_second": 190.352, | |
| "eval_steps_per_second": 24.337, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 3.194444444444444e-05, | |
| "loss": 0.2399, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.237, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "eval_accuracy": 0.7625570776255708, | |
| "eval_loss": 0.6790792942047119, | |
| "eval_runtime": 1.21, | |
| "eval_samples_per_second": 180.993, | |
| "eval_steps_per_second": 23.141, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 0.1989, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2505, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7625570776255708, | |
| "eval_loss": 0.5550326108932495, | |
| "eval_runtime": 1.0754, | |
| "eval_samples_per_second": 203.65, | |
| "eval_steps_per_second": 26.037, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "learning_rate": 2.6388888888888892e-05, | |
| "loss": 0.2188, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 10.73, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2018, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "eval_accuracy": 0.7625570776255708, | |
| "eval_loss": 0.5445629358291626, | |
| "eval_runtime": 1.0807, | |
| "eval_samples_per_second": 202.648, | |
| "eval_steps_per_second": 25.909, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 11.22, | |
| "learning_rate": 2.361111111111111e-05, | |
| "loss": 0.2028, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.71, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1841, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7442922374429224, | |
| "eval_loss": 0.5496687889099121, | |
| "eval_runtime": 1.1278, | |
| "eval_samples_per_second": 194.175, | |
| "eval_steps_per_second": 24.826, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 12.2, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.165, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.68, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.1692, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.98, | |
| "eval_accuracy": 0.771689497716895, | |
| "eval_loss": 0.5916743278503418, | |
| "eval_runtime": 1.1094, | |
| "eval_samples_per_second": 197.411, | |
| "eval_steps_per_second": 25.24, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 1.8055555555555555e-05, | |
| "loss": 0.2192, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.1624, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.776255707762557, | |
| "eval_loss": 0.5254391431808472, | |
| "eval_runtime": 1.1036, | |
| "eval_samples_per_second": 198.45, | |
| "eval_steps_per_second": 25.373, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 1.527777777777778e-05, | |
| "loss": 0.2206, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.1518, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "eval_accuracy": 0.7808219178082192, | |
| "eval_loss": 0.5295854806900024, | |
| "eval_runtime": 1.059, | |
| "eval_samples_per_second": 206.808, | |
| "eval_steps_per_second": 26.441, | |
| "step": 307 | |
| }, | |
| { | |
| "epoch": 15.12, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1465, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 15.61, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.1275, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7625570776255708, | |
| "eval_loss": 0.585834801197052, | |
| "eval_runtime": 1.2635, | |
| "eval_samples_per_second": 173.334, | |
| "eval_steps_per_second": 22.161, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 16.1, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.1237, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 16.59, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.1107, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "eval_accuracy": 0.776255707762557, | |
| "eval_loss": 0.591891348361969, | |
| "eval_runtime": 1.0639, | |
| "eval_samples_per_second": 205.84, | |
| "eval_steps_per_second": 26.317, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 17.07, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.1422, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 17.56, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1192, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.771689497716895, | |
| "eval_loss": 0.6027428507804871, | |
| "eval_runtime": 1.0536, | |
| "eval_samples_per_second": 207.851, | |
| "eval_steps_per_second": 26.575, | |
| "step": 369 | |
| }, | |
| { | |
| "epoch": 18.05, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.1806, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 18.54, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0842, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_accuracy": 0.771689497716895, | |
| "eval_loss": 0.6435291171073914, | |
| "eval_runtime": 1.0647, | |
| "eval_samples_per_second": 205.689, | |
| "eval_steps_per_second": 26.298, | |
| "step": 389 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "learning_rate": 1.388888888888889e-06, | |
| "loss": 0.1594, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 19.51, | |
| "learning_rate": 0.0, | |
| "loss": 0.1472, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.51, | |
| "eval_accuracy": 0.7671232876712328, | |
| "eval_loss": 0.6201618313789368, | |
| "eval_runtime": 1.0892, | |
| "eval_samples_per_second": 201.069, | |
| "eval_steps_per_second": 25.707, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.51, | |
| "step": 400, | |
| "total_flos": 2.493693637268521e+17, | |
| "train_loss": 0.3432186938822269, | |
| "train_runtime": 208.3396, | |
| "train_samples_per_second": 62.686, | |
| "train_steps_per_second": 1.92 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 400, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 2.493693637268521e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |