| { | |
| "best_metric": 1.0, | |
| "best_model_checkpoint": "videomae-base-finetuned-ssv2-finetuned-traffic-dataset-mae/checkpoint-168", | |
| "epoch": 7.011160714285714, | |
| "eval_steps": 500, | |
| "global_step": 397, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 14.671374320983887, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.7193, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 7.1641387939453125, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.5298, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 3.540851354598999, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.3106, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 1.9156169891357422, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.8657, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 2.8788669109344482, | |
| "learning_rate": 4.937965260545906e-05, | |
| "loss": 0.6144, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.27961549162864685, | |
| "eval_runtime": 24.2905, | |
| "eval_samples_per_second": 0.576, | |
| "eval_steps_per_second": 0.288, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 16.718944549560547, | |
| "learning_rate": 4.8138957816377175e-05, | |
| "loss": 0.4883, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 0.7438091039657593, | |
| "learning_rate": 4.689826302729529e-05, | |
| "loss": 0.236, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 1.806220293045044, | |
| "learning_rate": 4.56575682382134e-05, | |
| "loss": 0.4257, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 4.333924293518066, | |
| "learning_rate": 4.441687344913151e-05, | |
| "loss": 0.169, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 3.9030086994171143, | |
| "learning_rate": 4.317617866004963e-05, | |
| "loss": 0.313, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.010061407461762428, | |
| "learning_rate": 4.1935483870967746e-05, | |
| "loss": 0.0637, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_accuracy": 0.9285714285714286, | |
| "eval_loss": 0.11744929850101471, | |
| "eval_runtime": 24.0323, | |
| "eval_samples_per_second": 0.583, | |
| "eval_steps_per_second": 0.291, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 0.009479613974690437, | |
| "learning_rate": 4.069478908188586e-05, | |
| "loss": 0.2538, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.21475882828235626, | |
| "learning_rate": 3.945409429280397e-05, | |
| "loss": 0.1613, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 0.05664811283349991, | |
| "learning_rate": 3.8213399503722084e-05, | |
| "loss": 0.5442, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 1.2033185958862305, | |
| "learning_rate": 3.69727047146402e-05, | |
| "loss": 0.737, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 0.01298306230455637, | |
| "learning_rate": 3.573200992555831e-05, | |
| "loss": 0.4911, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.028704656288027763, | |
| "eval_runtime": 20.2796, | |
| "eval_samples_per_second": 0.69, | |
| "eval_steps_per_second": 0.345, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.12567199766635895, | |
| "learning_rate": 3.449131513647643e-05, | |
| "loss": 0.267, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 0.3274080753326416, | |
| "learning_rate": 3.325062034739454e-05, | |
| "loss": 0.2982, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 5.183434963226318, | |
| "learning_rate": 3.200992555831266e-05, | |
| "loss": 0.2744, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 0.20806659758090973, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.0124, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 0.04709279537200928, | |
| "learning_rate": 2.9528535980148887e-05, | |
| "loss": 0.0025, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 0.0054036942310631275, | |
| "learning_rate": 2.8287841191067e-05, | |
| "loss": 0.3625, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.08213090896606445, | |
| "eval_runtime": 24.8422, | |
| "eval_samples_per_second": 0.564, | |
| "eval_steps_per_second": 0.282, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 0.09801805019378662, | |
| "learning_rate": 2.7047146401985113e-05, | |
| "loss": 0.0082, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "grad_norm": 0.009371934458613396, | |
| "learning_rate": 2.5806451612903226e-05, | |
| "loss": 0.2331, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "grad_norm": 0.33655962347984314, | |
| "learning_rate": 2.4565756823821338e-05, | |
| "loss": 0.1461, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 0.009524806402623653, | |
| "learning_rate": 2.3325062034739454e-05, | |
| "loss": 0.3265, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "grad_norm": 0.0037162320222705603, | |
| "learning_rate": 2.208436724565757e-05, | |
| "loss": 0.001, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "grad_norm": 0.005654833745211363, | |
| "learning_rate": 2.0843672456575683e-05, | |
| "loss": 0.1472, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.27267220616340637, | |
| "eval_runtime": 24.1835, | |
| "eval_samples_per_second": 0.579, | |
| "eval_steps_per_second": 0.289, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "grad_norm": 1.0853424072265625, | |
| "learning_rate": 1.9602977667493796e-05, | |
| "loss": 0.0061, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "grad_norm": 0.10893365740776062, | |
| "learning_rate": 1.8362282878411912e-05, | |
| "loss": 0.0329, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "grad_norm": 0.003927825018763542, | |
| "learning_rate": 1.7121588089330025e-05, | |
| "loss": 0.0124, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "grad_norm": 0.03731616213917732, | |
| "learning_rate": 1.588089330024814e-05, | |
| "loss": 0.1577, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "grad_norm": 0.027107276022434235, | |
| "learning_rate": 1.4640198511166252e-05, | |
| "loss": 0.0068, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "eval_accuracy": 0.7142857142857143, | |
| "eval_loss": 1.524235486984253, | |
| "eval_runtime": 22.5747, | |
| "eval_samples_per_second": 0.62, | |
| "eval_steps_per_second": 0.31, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "grad_norm": 0.015361560508608818, | |
| "learning_rate": 1.3399503722084367e-05, | |
| "loss": 0.0155, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "grad_norm": 0.009546751156449318, | |
| "learning_rate": 1.2158808933002481e-05, | |
| "loss": 0.2251, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "grad_norm": 0.00286727212369442, | |
| "learning_rate": 1.0918114143920596e-05, | |
| "loss": 0.0003, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "grad_norm": 0.006020919419825077, | |
| "learning_rate": 9.67741935483871e-06, | |
| "loss": 0.0016, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "grad_norm": 0.0052633825689554214, | |
| "learning_rate": 8.436724565756825e-06, | |
| "loss": 0.2976, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "grad_norm": 0.006017320789396763, | |
| "learning_rate": 7.1960297766749385e-06, | |
| "loss": 0.0009, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.7878251671791077, | |
| "eval_runtime": 24.9105, | |
| "eval_samples_per_second": 0.562, | |
| "eval_steps_per_second": 0.281, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.026988975703716278, | |
| "eval_runtime": 5.8255, | |
| "eval_samples_per_second": 0.687, | |
| "eval_steps_per_second": 0.343, | |
| "step": 397 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.026988975703716278, | |
| "eval_runtime": 5.8152, | |
| "eval_samples_per_second": 0.688, | |
| "eval_steps_per_second": 0.344, | |
| "step": 397 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 448, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 9.769146381458473e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |