| { |
| "best_metric": 0.5004622936248779, |
| "best_model_checkpoint": "./vit-huge-augmented-ph2-patch-14/checkpoint-50", |
| "epoch": 4.0, |
| "global_step": 552, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.07, |
| "learning_rate": 0.00019637681159420293, |
| "loss": 0.4185, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.14, |
| "learning_rate": 0.0001927536231884058, |
| "loss": 0.1809, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.22, |
| "learning_rate": 0.0001891304347826087, |
| "loss": 0.1261, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.29, |
| "learning_rate": 0.0001855072463768116, |
| "loss": 0.0504, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.36, |
| "learning_rate": 0.0001818840579710145, |
| "loss": 0.0835, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.36, |
| "eval_accuracy": 0.8581196581196581, |
| "eval_f1": 0.8581196581196581, |
| "eval_loss": 0.5004622936248779, |
| "eval_precision": 0.8581196581196581, |
| "eval_recall": 0.8581196581196581, |
| "eval_runtime": 12.5476, |
| "eval_samples_per_second": 46.623, |
| "eval_steps_per_second": 5.898, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.43, |
| "learning_rate": 0.0001782608695652174, |
| "loss": 0.0372, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.51, |
| "learning_rate": 0.0001746376811594203, |
| "loss": 0.0741, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.58, |
| "learning_rate": 0.0001710144927536232, |
| "loss": 0.0406, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.65, |
| "learning_rate": 0.0001673913043478261, |
| "loss": 0.0156, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.72, |
| "learning_rate": 0.000163768115942029, |
| "loss": 0.0168, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.72, |
| "eval_accuracy": 0.8564102564102564, |
| "eval_f1": 0.8564102564102564, |
| "eval_loss": 0.5818132162094116, |
| "eval_precision": 0.8564102564102564, |
| "eval_recall": 0.8564102564102564, |
| "eval_runtime": 12.6681, |
| "eval_samples_per_second": 46.179, |
| "eval_steps_per_second": 5.841, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.8, |
| "learning_rate": 0.00016014492753623188, |
| "loss": 0.039, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.87, |
| "learning_rate": 0.0001565217391304348, |
| "loss": 0.0577, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.94, |
| "learning_rate": 0.00015289855072463769, |
| "loss": 0.0093, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.01, |
| "learning_rate": 0.00014927536231884058, |
| "loss": 0.0343, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.09, |
| "learning_rate": 0.0001456521739130435, |
| "loss": 0.0205, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.09, |
| "eval_accuracy": 0.8478632478632478, |
| "eval_f1": 0.8478632478632478, |
| "eval_loss": 0.5219599604606628, |
| "eval_precision": 0.8478632478632478, |
| "eval_recall": 0.8478632478632478, |
| "eval_runtime": 12.5714, |
| "eval_samples_per_second": 46.534, |
| "eval_steps_per_second": 5.886, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.16, |
| "learning_rate": 0.00014202898550724638, |
| "loss": 0.0124, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.23, |
| "learning_rate": 0.00013840579710144927, |
| "loss": 0.0048, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.3, |
| "learning_rate": 0.0001347826086956522, |
| "loss": 0.0042, |
| "step": 180 |
| }, |
| { |
| "epoch": 1.38, |
| "learning_rate": 0.00013115942028985508, |
| "loss": 0.0278, |
| "step": 190 |
| }, |
| { |
| "epoch": 1.45, |
| "learning_rate": 0.00012753623188405797, |
| "loss": 0.0125, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.45, |
| "eval_accuracy": 0.8512820512820513, |
| "eval_f1": 0.8512820512820513, |
| "eval_loss": 0.6188096404075623, |
| "eval_precision": 0.8512820512820513, |
| "eval_recall": 0.8512820512820513, |
| "eval_runtime": 12.5876, |
| "eval_samples_per_second": 46.474, |
| "eval_steps_per_second": 5.879, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.52, |
| "learning_rate": 0.00012391304347826086, |
| "loss": 0.0034, |
| "step": 210 |
| }, |
| { |
| "epoch": 1.59, |
| "learning_rate": 0.00012028985507246378, |
| "loss": 0.0034, |
| "step": 220 |
| }, |
| { |
| "epoch": 1.67, |
| "learning_rate": 0.00011666666666666668, |
| "loss": 0.0033, |
| "step": 230 |
| }, |
| { |
| "epoch": 1.74, |
| "learning_rate": 0.00011304347826086956, |
| "loss": 0.0029, |
| "step": 240 |
| }, |
| { |
| "epoch": 1.81, |
| "learning_rate": 0.00010942028985507247, |
| "loss": 0.0029, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.81, |
| "eval_accuracy": 0.8769230769230769, |
| "eval_f1": 0.8769230769230769, |
| "eval_loss": 0.6046304106712341, |
| "eval_precision": 0.8769230769230769, |
| "eval_recall": 0.8769230769230769, |
| "eval_runtime": 12.6151, |
| "eval_samples_per_second": 46.373, |
| "eval_steps_per_second": 5.866, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.88, |
| "learning_rate": 0.00010579710144927538, |
| "loss": 0.0027, |
| "step": 260 |
| }, |
| { |
| "epoch": 1.96, |
| "learning_rate": 0.00010217391304347828, |
| "loss": 0.0024, |
| "step": 270 |
| }, |
| { |
| "epoch": 2.03, |
| "learning_rate": 9.855072463768117e-05, |
| "loss": 0.0024, |
| "step": 280 |
| }, |
| { |
| "epoch": 2.1, |
| "learning_rate": 9.492753623188406e-05, |
| "loss": 0.0022, |
| "step": 290 |
| }, |
| { |
| "epoch": 2.17, |
| "learning_rate": 9.130434782608696e-05, |
| "loss": 0.0021, |
| "step": 300 |
| }, |
| { |
| "epoch": 2.17, |
| "eval_accuracy": 0.8803418803418803, |
| "eval_f1": 0.8803418803418802, |
| "eval_loss": 0.6407693028450012, |
| "eval_precision": 0.8803418803418803, |
| "eval_recall": 0.8803418803418803, |
| "eval_runtime": 12.5666, |
| "eval_samples_per_second": 46.552, |
| "eval_steps_per_second": 5.889, |
| "step": 300 |
| }, |
| { |
| "epoch": 2.25, |
| "learning_rate": 8.768115942028987e-05, |
| "loss": 0.0021, |
| "step": 310 |
| }, |
| { |
| "epoch": 2.32, |
| "learning_rate": 8.405797101449276e-05, |
| "loss": 0.002, |
| "step": 320 |
| }, |
| { |
| "epoch": 2.39, |
| "learning_rate": 8.043478260869566e-05, |
| "loss": 0.0019, |
| "step": 330 |
| }, |
| { |
| "epoch": 2.46, |
| "learning_rate": 7.681159420289855e-05, |
| "loss": 0.0018, |
| "step": 340 |
| }, |
| { |
| "epoch": 2.54, |
| "learning_rate": 7.318840579710145e-05, |
| "loss": 0.0018, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.54, |
| "eval_accuracy": 0.8803418803418803, |
| "eval_f1": 0.8803418803418802, |
| "eval_loss": 0.6587864756584167, |
| "eval_precision": 0.8803418803418803, |
| "eval_recall": 0.8803418803418803, |
| "eval_runtime": 12.6386, |
| "eval_samples_per_second": 46.287, |
| "eval_steps_per_second": 5.855, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.61, |
| "learning_rate": 6.956521739130436e-05, |
| "loss": 0.0017, |
| "step": 360 |
| }, |
| { |
| "epoch": 2.68, |
| "learning_rate": 6.594202898550726e-05, |
| "loss": 0.0017, |
| "step": 370 |
| }, |
| { |
| "epoch": 2.75, |
| "learning_rate": 6.231884057971015e-05, |
| "loss": 0.0017, |
| "step": 380 |
| }, |
| { |
| "epoch": 2.83, |
| "learning_rate": 5.869565217391305e-05, |
| "loss": 0.0016, |
| "step": 390 |
| }, |
| { |
| "epoch": 2.9, |
| "learning_rate": 5.507246376811594e-05, |
| "loss": 0.0015, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.9, |
| "eval_accuracy": 0.8803418803418803, |
| "eval_f1": 0.8803418803418802, |
| "eval_loss": 0.6720498204231262, |
| "eval_precision": 0.8803418803418803, |
| "eval_recall": 0.8803418803418803, |
| "eval_runtime": 12.6206, |
| "eval_samples_per_second": 46.353, |
| "eval_steps_per_second": 5.863, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.97, |
| "learning_rate": 5.144927536231884e-05, |
| "loss": 0.0016, |
| "step": 410 |
| }, |
| { |
| "epoch": 3.04, |
| "learning_rate": 4.782608695652174e-05, |
| "loss": 0.0015, |
| "step": 420 |
| }, |
| { |
| "epoch": 3.12, |
| "learning_rate": 4.4202898550724645e-05, |
| "loss": 0.0015, |
| "step": 430 |
| }, |
| { |
| "epoch": 3.19, |
| "learning_rate": 4.057971014492754e-05, |
| "loss": 0.0016, |
| "step": 440 |
| }, |
| { |
| "epoch": 3.26, |
| "learning_rate": 3.695652173913043e-05, |
| "loss": 0.0014, |
| "step": 450 |
| }, |
| { |
| "epoch": 3.26, |
| "eval_accuracy": 0.8803418803418803, |
| "eval_f1": 0.8803418803418802, |
| "eval_loss": 0.6803039312362671, |
| "eval_precision": 0.8803418803418803, |
| "eval_recall": 0.8803418803418803, |
| "eval_runtime": 12.6282, |
| "eval_samples_per_second": 46.325, |
| "eval_steps_per_second": 5.86, |
| "step": 450 |
| }, |
| { |
| "epoch": 3.33, |
| "learning_rate": 3.3333333333333335e-05, |
| "loss": 0.0015, |
| "step": 460 |
| }, |
| { |
| "epoch": 3.41, |
| "learning_rate": 2.971014492753623e-05, |
| "loss": 0.0014, |
| "step": 470 |
| }, |
| { |
| "epoch": 3.48, |
| "learning_rate": 2.608695652173913e-05, |
| "loss": 0.0014, |
| "step": 480 |
| }, |
| { |
| "epoch": 3.55, |
| "learning_rate": 2.246376811594203e-05, |
| "loss": 0.0014, |
| "step": 490 |
| }, |
| { |
| "epoch": 3.62, |
| "learning_rate": 1.8840579710144928e-05, |
| "loss": 0.0014, |
| "step": 500 |
| }, |
| { |
| "epoch": 3.62, |
| "eval_accuracy": 0.8803418803418803, |
| "eval_f1": 0.8803418803418802, |
| "eval_loss": 0.6861467361450195, |
| "eval_precision": 0.8803418803418803, |
| "eval_recall": 0.8803418803418803, |
| "eval_runtime": 12.7312, |
| "eval_samples_per_second": 45.95, |
| "eval_steps_per_second": 5.812, |
| "step": 500 |
| }, |
| { |
| "epoch": 3.7, |
| "learning_rate": 1.5217391304347828e-05, |
| "loss": 0.0014, |
| "step": 510 |
| }, |
| { |
| "epoch": 3.77, |
| "learning_rate": 1.1594202898550725e-05, |
| "loss": 0.0014, |
| "step": 520 |
| }, |
| { |
| "epoch": 3.84, |
| "learning_rate": 7.971014492753623e-06, |
| "loss": 0.0013, |
| "step": 530 |
| }, |
| { |
| "epoch": 3.91, |
| "learning_rate": 4.347826086956522e-06, |
| "loss": 0.0014, |
| "step": 540 |
| }, |
| { |
| "epoch": 3.99, |
| "learning_rate": 7.246376811594203e-07, |
| "loss": 0.0014, |
| "step": 550 |
| }, |
| { |
| "epoch": 3.99, |
| "eval_accuracy": 0.8803418803418803, |
| "eval_f1": 0.8803418803418802, |
| "eval_loss": 0.6879416108131409, |
| "eval_precision": 0.8803418803418803, |
| "eval_recall": 0.8803418803418803, |
| "eval_runtime": 12.5938, |
| "eval_samples_per_second": 46.451, |
| "eval_steps_per_second": 5.876, |
| "step": 550 |
| }, |
| { |
| "epoch": 4.0, |
| "step": 552, |
| "total_flos": 5.013262291864781e+18, |
| "train_loss": 0.024155851723491283, |
| "train_runtime": 1143.7087, |
| "train_samples_per_second": 7.694, |
| "train_steps_per_second": 0.483 |
| } |
| ], |
| "max_steps": 552, |
| "num_train_epochs": 4, |
| "total_flos": 5.013262291864781e+18, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
