| { | |
| "best_metric": 0.44582676887512207, | |
| "best_model_checkpoint": "dit-base-finetuned-rvlcdip-finetuned-mobile-eye-tracking-dataset-v2/checkpoint-198", | |
| "epoch": 30.0, | |
| "eval_steps": 500, | |
| "global_step": 270, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3671875, | |
| "eval_loss": 1.3608635663986206, | |
| "eval_runtime": 1.1166, | |
| "eval_samples_per_second": 114.632, | |
| "eval_steps_per_second": 3.582, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 1.3865, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6484375, | |
| "eval_loss": 1.234185814857483, | |
| "eval_runtime": 1.092, | |
| "eval_samples_per_second": 117.221, | |
| "eval_steps_per_second": 3.663, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 1.3053, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.703125, | |
| "eval_loss": 1.110687494277954, | |
| "eval_runtime": 1.0603, | |
| "eval_samples_per_second": 120.722, | |
| "eval_steps_per_second": 3.773, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 4.938271604938271e-05, | |
| "loss": 1.1777, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6953125, | |
| "eval_loss": 1.0204391479492188, | |
| "eval_runtime": 1.2761, | |
| "eval_samples_per_second": 100.304, | |
| "eval_steps_per_second": 3.135, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 4.732510288065844e-05, | |
| "loss": 1.0704, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.703125, | |
| "eval_loss": 0.8894159197807312, | |
| "eval_runtime": 0.9836, | |
| "eval_samples_per_second": 130.137, | |
| "eval_steps_per_second": 4.067, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 4.5267489711934157e-05, | |
| "loss": 0.9475, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6953125, | |
| "eval_loss": 0.7666884660720825, | |
| "eval_runtime": 1.5089, | |
| "eval_samples_per_second": 84.828, | |
| "eval_steps_per_second": 2.651, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 4.3209876543209875e-05, | |
| "loss": 0.8477, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7265625, | |
| "eval_loss": 0.720241367816925, | |
| "eval_runtime": 1.2244, | |
| "eval_samples_per_second": 104.542, | |
| "eval_steps_per_second": 3.267, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 4.11522633744856e-05, | |
| "loss": 0.7669, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.734375, | |
| "eval_loss": 0.6863818168640137, | |
| "eval_runtime": 1.3256, | |
| "eval_samples_per_second": 96.562, | |
| "eval_steps_per_second": 3.018, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 3.909465020576132e-05, | |
| "loss": 0.7125, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8203125, | |
| "eval_loss": 0.5822394490242004, | |
| "eval_runtime": 1.3032, | |
| "eval_samples_per_second": 98.223, | |
| "eval_steps_per_second": 3.069, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.6915, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.765625, | |
| "eval_loss": 0.6075575351715088, | |
| "eval_runtime": 0.976, | |
| "eval_samples_per_second": 131.154, | |
| "eval_steps_per_second": 4.099, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.765625, | |
| "eval_loss": 0.5758803486824036, | |
| "eval_runtime": 0.9718, | |
| "eval_samples_per_second": 131.708, | |
| "eval_steps_per_second": 4.116, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 3.497942386831276e-05, | |
| "loss": 0.6984, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.78125, | |
| "eval_loss": 0.5367965698242188, | |
| "eval_runtime": 1.1194, | |
| "eval_samples_per_second": 114.35, | |
| "eval_steps_per_second": 3.573, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "learning_rate": 3.292181069958848e-05, | |
| "loss": 0.6443, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.84375, | |
| "eval_loss": 0.5282293558120728, | |
| "eval_runtime": 0.91, | |
| "eval_samples_per_second": 140.655, | |
| "eval_steps_per_second": 4.395, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 3.08641975308642e-05, | |
| "loss": 0.6293, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.5335959792137146, | |
| "eval_runtime": 1.0654, | |
| "eval_samples_per_second": 120.14, | |
| "eval_steps_per_second": 3.754, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 14.44, | |
| "learning_rate": 2.880658436213992e-05, | |
| "loss": 0.6411, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.7265625, | |
| "eval_loss": 0.5863245129585266, | |
| "eval_runtime": 1.2348, | |
| "eval_samples_per_second": 103.66, | |
| "eval_steps_per_second": 3.239, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 15.56, | |
| "learning_rate": 2.6748971193415638e-05, | |
| "loss": 0.6126, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.78125, | |
| "eval_loss": 0.5172436237335205, | |
| "eval_runtime": 0.975, | |
| "eval_samples_per_second": 131.285, | |
| "eval_steps_per_second": 4.103, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 2.4691358024691357e-05, | |
| "loss": 0.603, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8046875, | |
| "eval_loss": 0.5282721519470215, | |
| "eval_runtime": 0.9219, | |
| "eval_samples_per_second": 138.844, | |
| "eval_steps_per_second": 4.339, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 17.78, | |
| "learning_rate": 2.2633744855967078e-05, | |
| "loss": 0.6065, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.78125, | |
| "eval_loss": 0.4975610375404358, | |
| "eval_runtime": 1.0828, | |
| "eval_samples_per_second": 118.21, | |
| "eval_steps_per_second": 3.694, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 18.89, | |
| "learning_rate": 2.05761316872428e-05, | |
| "loss": 0.5757, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.78125, | |
| "eval_loss": 0.5058664679527283, | |
| "eval_runtime": 1.0696, | |
| "eval_samples_per_second": 119.668, | |
| "eval_steps_per_second": 3.74, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.5521, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8203125, | |
| "eval_loss": 0.47192054986953735, | |
| "eval_runtime": 1.1963, | |
| "eval_samples_per_second": 106.997, | |
| "eval_steps_per_second": 3.344, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.78125, | |
| "eval_loss": 0.4766509234905243, | |
| "eval_runtime": 0.7066, | |
| "eval_samples_per_second": 181.139, | |
| "eval_steps_per_second": 5.661, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 21.11, | |
| "learning_rate": 1.646090534979424e-05, | |
| "loss": 0.5548, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8359375, | |
| "eval_loss": 0.44582676887512207, | |
| "eval_runtime": 1.042, | |
| "eval_samples_per_second": 122.836, | |
| "eval_steps_per_second": 3.839, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "learning_rate": 1.440329218106996e-05, | |
| "loss": 0.5368, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.8046875, | |
| "eval_loss": 0.46391141414642334, | |
| "eval_runtime": 0.9999, | |
| "eval_samples_per_second": 128.014, | |
| "eval_steps_per_second": 4.0, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 23.33, | |
| "learning_rate": 1.2345679012345678e-05, | |
| "loss": 0.5295, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.828125, | |
| "eval_loss": 0.45563599467277527, | |
| "eval_runtime": 1.037, | |
| "eval_samples_per_second": 123.433, | |
| "eval_steps_per_second": 3.857, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 24.44, | |
| "learning_rate": 1.02880658436214e-05, | |
| "loss": 0.5474, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.796875, | |
| "eval_loss": 0.4770171344280243, | |
| "eval_runtime": 0.9465, | |
| "eval_samples_per_second": 135.233, | |
| "eval_steps_per_second": 4.226, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 25.56, | |
| "learning_rate": 8.23045267489712e-06, | |
| "loss": 0.5053, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8203125, | |
| "eval_loss": 0.4503433108329773, | |
| "eval_runtime": 0.9843, | |
| "eval_samples_per_second": 130.045, | |
| "eval_steps_per_second": 4.064, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "learning_rate": 6.172839506172839e-06, | |
| "loss": 0.5076, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.796875, | |
| "eval_loss": 0.457803875207901, | |
| "eval_runtime": 1.2313, | |
| "eval_samples_per_second": 103.957, | |
| "eval_steps_per_second": 3.249, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 27.78, | |
| "learning_rate": 4.11522633744856e-06, | |
| "loss": 0.5058, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7890625, | |
| "eval_loss": 0.47135162353515625, | |
| "eval_runtime": 1.2208, | |
| "eval_samples_per_second": 104.851, | |
| "eval_steps_per_second": 3.277, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 28.89, | |
| "learning_rate": 2.05761316872428e-06, | |
| "loss": 0.4792, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.8046875, | |
| "eval_loss": 0.4589694142341614, | |
| "eval_runtime": 0.8424, | |
| "eval_samples_per_second": 151.955, | |
| "eval_steps_per_second": 4.749, | |
| "step": 261 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.494, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8046875, | |
| "eval_loss": 0.45556190609931946, | |
| "eval_runtime": 0.9885, | |
| "eval_samples_per_second": 129.491, | |
| "eval_steps_per_second": 4.047, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 270, | |
| "total_flos": 2.678458830909604e+18, | |
| "train_loss": 0.70849390912939, | |
| "train_runtime": 653.9668, | |
| "train_samples_per_second": 52.847, | |
| "train_steps_per_second": 0.413 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 270, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 2.678458830909604e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |