| { | |
| "best_metric": 0.5443037974683544, | |
| "best_model_checkpoint": "videomae-base-finetuned-kinetics-finetuned-caer-final-4-epochs\\checkpoint-246", | |
| "epoch": 7.017123287671233, | |
| "eval_steps": 500, | |
| "global_step": 584, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 6.458174705505371, | |
| "learning_rate": 8.47457627118644e-06, | |
| "loss": 1.602, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 7.2191619873046875, | |
| "learning_rate": 1.694915254237288e-05, | |
| "loss": 1.6082, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 6.413382053375244, | |
| "learning_rate": 2.5423728813559322e-05, | |
| "loss": 1.6091, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 6.1725263595581055, | |
| "learning_rate": 3.389830508474576e-05, | |
| "loss": 1.5506, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 6.784051418304443, | |
| "learning_rate": 4.2372881355932206e-05, | |
| "loss": 1.5818, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 6.475350856781006, | |
| "learning_rate": 4.990476190476191e-05, | |
| "loss": 1.5237, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 8.449509620666504, | |
| "learning_rate": 4.895238095238096e-05, | |
| "loss": 1.6543, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 5.603182792663574, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.3388, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_accuracy": 0.3924050632911392, | |
| "eval_loss": 1.4801665544509888, | |
| "eval_runtime": 25.4554, | |
| "eval_samples_per_second": 3.103, | |
| "eval_steps_per_second": 0.55, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 8.551097869873047, | |
| "learning_rate": 4.704761904761905e-05, | |
| "loss": 1.4052, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 11.188709259033203, | |
| "learning_rate": 4.60952380952381e-05, | |
| "loss": 1.345, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 9.687131881713867, | |
| "learning_rate": 4.514285714285714e-05, | |
| "loss": 1.3573, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 10.117310523986816, | |
| "learning_rate": 4.419047619047619e-05, | |
| "loss": 1.1564, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 7.854645729064941, | |
| "learning_rate": 4.323809523809524e-05, | |
| "loss": 1.2247, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 7.707760334014893, | |
| "learning_rate": 4.228571428571429e-05, | |
| "loss": 1.1877, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 8.141950607299805, | |
| "learning_rate": 4.133333333333333e-05, | |
| "loss": 1.0836, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 9.751481056213379, | |
| "learning_rate": 4.038095238095238e-05, | |
| "loss": 1.1881, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_accuracy": 0.3670886075949367, | |
| "eval_loss": 1.3925659656524658, | |
| "eval_runtime": 25.2241, | |
| "eval_samples_per_second": 3.132, | |
| "eval_steps_per_second": 0.555, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 13.817303657531738, | |
| "learning_rate": 3.942857142857143e-05, | |
| "loss": 0.8556, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 12.339946746826172, | |
| "learning_rate": 3.847619047619048e-05, | |
| "loss": 0.862, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 14.686485290527344, | |
| "learning_rate": 3.752380952380953e-05, | |
| "loss": 0.8869, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 9.216782569885254, | |
| "learning_rate": 3.6571428571428576e-05, | |
| "loss": 0.9421, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 12.398260116577148, | |
| "learning_rate": 3.561904761904762e-05, | |
| "loss": 0.695, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 11.731080055236816, | |
| "learning_rate": 3.466666666666667e-05, | |
| "loss": 1.0937, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 10.27041244506836, | |
| "learning_rate": 3.3714285714285716e-05, | |
| "loss": 0.9285, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 15.88553524017334, | |
| "learning_rate": 3.276190476190477e-05, | |
| "loss": 0.9722, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.5443037974683544, | |
| "eval_loss": 1.2269208431243896, | |
| "eval_runtime": 26.6893, | |
| "eval_samples_per_second": 2.96, | |
| "eval_steps_per_second": 0.525, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 10.362627029418945, | |
| "learning_rate": 3.180952380952381e-05, | |
| "loss": 0.9198, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 7.394880771636963, | |
| "learning_rate": 3.0857142857142856e-05, | |
| "loss": 0.5962, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 7.7396345138549805, | |
| "learning_rate": 2.9904761904761907e-05, | |
| "loss": 0.5537, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 15.393024444580078, | |
| "learning_rate": 2.8952380952380952e-05, | |
| "loss": 0.6155, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 13.79691219329834, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.7206, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 11.387887954711914, | |
| "learning_rate": 2.704761904761905e-05, | |
| "loss": 0.5823, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "grad_norm": 19.518312454223633, | |
| "learning_rate": 2.6095238095238095e-05, | |
| "loss": 0.7304, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 8.372328758239746, | |
| "learning_rate": 2.5142857142857147e-05, | |
| "loss": 0.6545, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "eval_accuracy": 0.4936708860759494, | |
| "eval_loss": 1.229438066482544, | |
| "eval_runtime": 27.627, | |
| "eval_samples_per_second": 2.86, | |
| "eval_steps_per_second": 0.507, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 9.884891510009766, | |
| "learning_rate": 2.419047619047619e-05, | |
| "loss": 0.6225, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "grad_norm": 24.8238525390625, | |
| "learning_rate": 2.323809523809524e-05, | |
| "loss": 0.4376, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "grad_norm": 11.315938949584961, | |
| "learning_rate": 2.2285714285714287e-05, | |
| "loss": 0.4116, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "grad_norm": 12.32546615600586, | |
| "learning_rate": 2.1333333333333335e-05, | |
| "loss": 0.4504, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "grad_norm": 7.664936065673828, | |
| "learning_rate": 2.0380952380952382e-05, | |
| "loss": 0.6834, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 16.222686767578125, | |
| "learning_rate": 1.942857142857143e-05, | |
| "loss": 0.5456, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "grad_norm": 7.430470943450928, | |
| "learning_rate": 1.8476190476190478e-05, | |
| "loss": 0.4284, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "grad_norm": 2.954035520553589, | |
| "learning_rate": 1.7523809523809526e-05, | |
| "loss": 0.4361, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "grad_norm": 27.79668426513672, | |
| "learning_rate": 1.657142857142857e-05, | |
| "loss": 0.5289, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "eval_accuracy": 0.4810126582278481, | |
| "eval_loss": 1.3317660093307495, | |
| "eval_runtime": 27.2349, | |
| "eval_samples_per_second": 2.901, | |
| "eval_steps_per_second": 0.514, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "grad_norm": 6.09324836730957, | |
| "learning_rate": 1.561904761904762e-05, | |
| "loss": 0.3316, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "grad_norm": 13.406198501586914, | |
| "learning_rate": 1.4666666666666668e-05, | |
| "loss": 0.2502, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "grad_norm": 10.042999267578125, | |
| "learning_rate": 1.3714285714285716e-05, | |
| "loss": 0.4386, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "grad_norm": 10.368085861206055, | |
| "learning_rate": 1.2761904761904764e-05, | |
| "loss": 0.3197, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "grad_norm": 14.689448356628418, | |
| "learning_rate": 1.180952380952381e-05, | |
| "loss": 0.4345, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "grad_norm": 5.982275009155273, | |
| "learning_rate": 1.0857142857142858e-05, | |
| "loss": 0.3263, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "grad_norm": 12.9437255859375, | |
| "learning_rate": 9.904761904761906e-06, | |
| "loss": 0.3721, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "grad_norm": 17.087923049926758, | |
| "learning_rate": 8.952380952380952e-06, | |
| "loss": 0.2826, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "eval_accuracy": 0.4810126582278481, | |
| "eval_loss": 1.254180908203125, | |
| "eval_runtime": 27.4561, | |
| "eval_samples_per_second": 2.877, | |
| "eval_steps_per_second": 0.51, | |
| "step": 492 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "grad_norm": 11.193989753723145, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.1963, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "grad_norm": 2.909717559814453, | |
| "learning_rate": 7.0476190476190475e-06, | |
| "loss": 0.1711, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "grad_norm": 14.048554420471191, | |
| "learning_rate": 6.095238095238095e-06, | |
| "loss": 0.1742, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "grad_norm": 14.992708206176758, | |
| "learning_rate": 5.142857142857143e-06, | |
| "loss": 0.3157, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "grad_norm": 13.224514961242676, | |
| "learning_rate": 4.19047619047619e-06, | |
| "loss": 0.2316, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "grad_norm": 20.14113998413086, | |
| "learning_rate": 3.238095238095238e-06, | |
| "loss": 0.1789, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "grad_norm": 5.539999008178711, | |
| "learning_rate": 2.285714285714286e-06, | |
| "loss": 0.1952, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "grad_norm": 1.642961025238037, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 0.2948, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "eval_accuracy": 0.5443037974683544, | |
| "eval_loss": 1.2192754745483398, | |
| "eval_runtime": 27.124, | |
| "eval_samples_per_second": 2.913, | |
| "eval_steps_per_second": 0.516, | |
| "step": 574 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "grad_norm": 1.9858059883117676, | |
| "learning_rate": 3.8095238095238096e-07, | |
| "loss": 0.1954, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "eval_accuracy": 0.5316455696202531, | |
| "eval_loss": 1.2165864706039429, | |
| "eval_runtime": 26.4535, | |
| "eval_samples_per_second": 2.986, | |
| "eval_steps_per_second": 0.529, | |
| "step": 584 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "step": 584, | |
| "total_flos": 4.331436394978345e+18, | |
| "train_loss": 0.7601832610695329, | |
| "train_runtime": 1891.5972, | |
| "train_samples_per_second": 1.852, | |
| "train_steps_per_second": 0.309 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "eval_accuracy": 0.4642857142857143, | |
| "eval_loss": 1.2583760023117065, | |
| "eval_runtime": 18.8662, | |
| "eval_samples_per_second": 2.968, | |
| "eval_steps_per_second": 0.53, | |
| "step": 584 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "eval_accuracy": 0.5443037974683544, | |
| "eval_loss": 1.2269208431243896, | |
| "eval_runtime": 27.3703, | |
| "eval_samples_per_second": 2.886, | |
| "eval_steps_per_second": 0.512, | |
| "step": 584 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 584, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 4.331436394978345e+18, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |