| { | |
| "best_metric": 0.8478260869565217, | |
| "best_model_checkpoint": "vit-base-patch16-224-ve-U12-b-24\\checkpoint-84", | |
| "epoch": 22.153846153846153, | |
| "eval_steps": 500, | |
| "global_step": 144, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.41304347826086957, | |
| "eval_loss": 1.3805949687957764, | |
| "eval_runtime": 0.7506, | |
| "eval_samples_per_second": 61.288, | |
| "eval_steps_per_second": 2.665, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 5.419117647058824e-05, | |
| "loss": 1.379, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5434782608695652, | |
| "eval_loss": 1.3102775812149048, | |
| "eval_runtime": 0.766, | |
| "eval_samples_per_second": 60.055, | |
| "eval_steps_per_second": 2.611, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_accuracy": 0.41304347826086957, | |
| "eval_loss": 1.2269459962844849, | |
| "eval_runtime": 0.7708, | |
| "eval_samples_per_second": 59.68, | |
| "eval_steps_per_second": 2.595, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 5.0147058823529414e-05, | |
| "loss": 1.2758, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.14116632938385, | |
| "eval_runtime": 0.7485, | |
| "eval_samples_per_second": 61.454, | |
| "eval_steps_per_second": 2.672, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 4.610294117647059e-05, | |
| "loss": 1.121, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "eval_accuracy": 0.4782608695652174, | |
| "eval_loss": 1.0649745464324951, | |
| "eval_runtime": 0.7498, | |
| "eval_samples_per_second": 61.347, | |
| "eval_steps_per_second": 2.667, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5217391304347826, | |
| "eval_loss": 1.0084419250488281, | |
| "eval_runtime": 0.7654, | |
| "eval_samples_per_second": 60.102, | |
| "eval_steps_per_second": 2.613, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 4.205882352941176e-05, | |
| "loss": 0.9871, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 0.9395221471786499, | |
| "eval_runtime": 0.7698, | |
| "eval_samples_per_second": 59.755, | |
| "eval_steps_per_second": 2.598, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 3.801470588235294e-05, | |
| "loss": 0.8612, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 0.8797907829284668, | |
| "eval_runtime": 0.7584, | |
| "eval_samples_per_second": 60.653, | |
| "eval_steps_per_second": 2.637, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 8.92, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 0.8218574523925781, | |
| "eval_runtime": 0.7333, | |
| "eval_samples_per_second": 62.731, | |
| "eval_steps_per_second": 2.727, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 9.23, | |
| "learning_rate": 3.397058823529412e-05, | |
| "loss": 0.7653, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.782608695652174, | |
| "eval_loss": 0.771176815032959, | |
| "eval_runtime": 0.7602, | |
| "eval_samples_per_second": 60.512, | |
| "eval_steps_per_second": 2.631, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "learning_rate": 2.992647058823529e-05, | |
| "loss": 0.6674, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 10.92, | |
| "eval_accuracy": 0.7608695652173914, | |
| "eval_loss": 0.7328380346298218, | |
| "eval_runtime": 0.7787, | |
| "eval_samples_per_second": 59.07, | |
| "eval_steps_per_second": 2.568, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 0.696773886680603, | |
| "eval_runtime": 0.7599, | |
| "eval_samples_per_second": 60.535, | |
| "eval_steps_per_second": 2.632, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "learning_rate": 2.5882352941176472e-05, | |
| "loss": 0.568, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 12.92, | |
| "eval_accuracy": 0.8478260869565217, | |
| "eval_loss": 0.645609974861145, | |
| "eval_runtime": 0.7242, | |
| "eval_samples_per_second": 63.52, | |
| "eval_steps_per_second": 2.762, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 13.85, | |
| "learning_rate": 2.1838235294117645e-05, | |
| "loss": 0.4723, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8043478260869565, | |
| "eval_loss": 0.6528236269950867, | |
| "eval_runtime": 0.758, | |
| "eval_samples_per_second": 60.687, | |
| "eval_steps_per_second": 2.639, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 14.92, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.7106918692588806, | |
| "eval_runtime": 0.7842, | |
| "eval_samples_per_second": 58.662, | |
| "eval_steps_per_second": 2.551, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 1.7794117647058825e-05, | |
| "loss": 0.4256, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7608695652173914, | |
| "eval_loss": 0.6335130929946899, | |
| "eval_runtime": 0.759, | |
| "eval_samples_per_second": 60.609, | |
| "eval_steps_per_second": 2.635, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 16.92, | |
| "learning_rate": 1.375e-05, | |
| "loss": 0.3524, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 16.92, | |
| "eval_accuracy": 0.8260869565217391, | |
| "eval_loss": 0.5952833294868469, | |
| "eval_runtime": 0.7509, | |
| "eval_samples_per_second": 61.258, | |
| "eval_steps_per_second": 2.663, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8260869565217391, | |
| "eval_loss": 0.5824078917503357, | |
| "eval_runtime": 0.7471, | |
| "eval_samples_per_second": 61.571, | |
| "eval_steps_per_second": 2.677, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 18.46, | |
| "learning_rate": 9.705882352941177e-06, | |
| "loss": 0.3282, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 18.92, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 0.6328944563865662, | |
| "eval_runtime": 0.7504, | |
| "eval_samples_per_second": 61.299, | |
| "eval_steps_per_second": 2.665, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 5.661764705882353e-06, | |
| "loss": 0.3074, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8043478260869565, | |
| "eval_loss": 0.5775049328804016, | |
| "eval_runtime": 0.7695, | |
| "eval_samples_per_second": 59.778, | |
| "eval_steps_per_second": 2.599, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 20.92, | |
| "eval_accuracy": 0.8043478260869565, | |
| "eval_loss": 0.5769753456115723, | |
| "eval_runtime": 0.7334, | |
| "eval_samples_per_second": 62.719, | |
| "eval_steps_per_second": 2.727, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 21.54, | |
| "learning_rate": 1.6176470588235295e-06, | |
| "loss": 0.3076, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8260869565217391, | |
| "eval_loss": 0.5748651623725891, | |
| "eval_runtime": 0.7702, | |
| "eval_samples_per_second": 59.723, | |
| "eval_steps_per_second": 2.597, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 22.15, | |
| "eval_accuracy": 0.8260869565217391, | |
| "eval_loss": 0.5747259855270386, | |
| "eval_runtime": 0.7582, | |
| "eval_samples_per_second": 60.667, | |
| "eval_steps_per_second": 2.638, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 22.15, | |
| "step": 144, | |
| "total_flos": 1.4061948496538665e+18, | |
| "train_loss": 0.6897724643349648, | |
| "train_runtime": 319.2612, | |
| "train_samples_per_second": 61.567, | |
| "train_steps_per_second": 0.451 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 144, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 24, | |
| "save_steps": 500, | |
| "total_flos": 1.4061948496538665e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |