| { | |
| "best_metric": 0.9285714285714286, | |
| "best_model_checkpoint": "timesformer-base-finetuned-k400-finetuned-kinetic-subset-three-local-temporal-with-spatial/checkpoint-150", | |
| "epoch": 5.166666666666667, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 32.2842903137207, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.6309, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 29.811689376831055, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.7024, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 26.413127899169922, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5057, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 17.640932083129883, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.8117, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 33.26347732543945, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.2937, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.8690476190476191, | |
| "eval_loss": 0.3884676396846771, | |
| "eval_runtime": 105.2382, | |
| "eval_samples_per_second": 1.596, | |
| "eval_steps_per_second": 1.596, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 0.8504588007926941, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.4217, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 0.1911085844039917, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.2901, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.12057169526815414, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.0256, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 0.2651258111000061, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.0627, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 0.1569805145263672, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2365, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_accuracy": 0.8809523809523809, | |
| "eval_loss": 0.600234866142273, | |
| "eval_runtime": 109.7974, | |
| "eval_samples_per_second": 1.53, | |
| "eval_steps_per_second": 1.53, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 0.038823891431093216, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.0043, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 0.037996720522642136, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0062, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 0.05053324997425079, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.0297, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 1.2805042266845703, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.0046, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 0.19171175360679626, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0164, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_accuracy": 0.9285714285714286, | |
| "eval_loss": 0.2311260849237442, | |
| "eval_runtime": 108.5839, | |
| "eval_samples_per_second": 1.547, | |
| "eval_steps_per_second": 1.547, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 0.025192299857735634, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.1078, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 0.02108108066022396, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.6885, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 0.008117140270769596, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0011, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 0.0680527612566948, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.0015, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "grad_norm": 0.011410138569772243, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.4057, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "eval_accuracy": 0.8630952380952381, | |
| "eval_loss": 0.5045905113220215, | |
| "eval_runtime": 109.1813, | |
| "eval_samples_per_second": 1.539, | |
| "eval_steps_per_second": 1.539, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 0.015822872519493103, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0012, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "grad_norm": 0.06866960972547531, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.0008, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "grad_norm": 0.025955336168408394, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.0014, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "grad_norm": 0.027819635346531868, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.034, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "grad_norm": 0.011835525743663311, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.0004, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "eval_accuracy": 0.875, | |
| "eval_loss": 0.4492361545562744, | |
| "eval_runtime": 108.7199, | |
| "eval_samples_per_second": 1.545, | |
| "eval_steps_per_second": 1.545, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "grad_norm": 0.9948074817657471, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.0022, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "grad_norm": 0.07322775572538376, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0011, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "grad_norm": 0.01132030040025711, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.0146, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "grad_norm": 48.047698974609375, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.0825, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "grad_norm": 0.02245236746966839, | |
| "learning_rate": 0.0, | |
| "loss": 0.0007, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "eval_accuracy": 0.8928571428571429, | |
| "eval_loss": 0.35668858885765076, | |
| "eval_runtime": 108.8318, | |
| "eval_samples_per_second": 1.544, | |
| "eval_steps_per_second": 1.544, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "step": 300, | |
| "total_flos": 3.28565653420032e+17, | |
| "train_loss": 0.2795241907471791, | |
| "train_runtime": 838.6117, | |
| "train_samples_per_second": 0.358, | |
| "train_steps_per_second": 0.358 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "eval_accuracy": 0.8987341772151899, | |
| "eval_loss": 0.49271300435066223, | |
| "eval_runtime": 66.5373, | |
| "eval_samples_per_second": 2.375, | |
| "eval_steps_per_second": 2.375, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "eval_accuracy": 0.8987341772151899, | |
| "eval_loss": 0.492713063955307, | |
| "eval_runtime": 65.6092, | |
| "eval_samples_per_second": 2.408, | |
| "eval_steps_per_second": 2.408, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "eval_accuracy": 0.8987341772151899, | |
| "eval_loss": 0.492713063955307, | |
| "eval_runtime": 66.3108, | |
| "eval_samples_per_second": 2.383, | |
| "eval_steps_per_second": 2.383, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 3.28565653420032e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |