timesformer-base-finetuned-k400-finetuned-kinetic400-subset-epoch6-num_frame_10
/
trainer_state.json
| { | |
| "best_metric": 0.8928571428571429, | |
| "best_model_checkpoint": "timesformer-base-finetuned-k400-finetuned-kinetic400-subset-epoch6-num_frame_10/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": 36.570594787597656, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.6089, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 22.93528175354004, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.3507, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 14.943221092224121, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0717, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 6.05604362487793, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.5376, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 18.73883819580078, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.204, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.8511904761904762, | |
| "eval_loss": 0.5931795835494995, | |
| "eval_runtime": 125.3965, | |
| "eval_samples_per_second": 1.34, | |
| "eval_steps_per_second": 1.34, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 0.47039151191711426, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.3619, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 0.3019910156726837, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.0952, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.09102782607078552, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.0109, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 1.194956660270691, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.0508, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 0.06759972870349884, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.3192, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_accuracy": 0.7738095238095238, | |
| "eval_loss": 1.044474482536316, | |
| "eval_runtime": 126.6326, | |
| "eval_samples_per_second": 1.327, | |
| "eval_steps_per_second": 1.327, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 0.16507962346076965, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.0055, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 0.09454397112131119, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1223, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 0.06931765377521515, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.0049, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 0.1607487052679062, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.0038, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 1.1081467866897583, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0072, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_accuracy": 0.8928571428571429, | |
| "eval_loss": 0.27960675954818726, | |
| "eval_runtime": 126.6296, | |
| "eval_samples_per_second": 1.327, | |
| "eval_steps_per_second": 1.327, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 0.019999559968709946, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.3039, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 0.022100064903497696, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.2629, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 0.008461786434054375, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.003, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 0.06302022933959961, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.0012, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "grad_norm": 0.008670525625348091, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.2034, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "eval_accuracy": 0.8690476190476191, | |
| "eval_loss": 0.6141949892044067, | |
| "eval_runtime": 126.8462, | |
| "eval_samples_per_second": 1.324, | |
| "eval_steps_per_second": 1.324, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 0.051909659057855606, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0009, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "grad_norm": 0.04402831196784973, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.0007, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "grad_norm": 0.029709341004490852, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.0258, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "grad_norm": 0.01658117026090622, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0909, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "grad_norm": 0.010167304426431656, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.0004, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "eval_accuracy": 0.8690476190476191, | |
| "eval_loss": 0.6565865278244019, | |
| "eval_runtime": 125.2366, | |
| "eval_samples_per_second": 1.341, | |
| "eval_steps_per_second": 1.341, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "grad_norm": 0.681441068649292, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.0018, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "grad_norm": 0.04809751361608505, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0005, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "grad_norm": 0.018213847652077675, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.0064, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "grad_norm": 59.33660888671875, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.0941, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "grad_norm": 0.014356265775859356, | |
| "learning_rate": 0.0, | |
| "loss": 0.0005, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "eval_accuracy": 0.8511904761904762, | |
| "eval_loss": 0.6626848578453064, | |
| "eval_runtime": 125.836, | |
| "eval_samples_per_second": 1.335, | |
| "eval_steps_per_second": 1.335, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "step": 300, | |
| "total_flos": 3.28565653420032e+17, | |
| "train_loss": 0.22503100648444765, | |
| "train_runtime": 990.4584, | |
| "train_samples_per_second": 0.303, | |
| "train_steps_per_second": 0.303 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "eval_accuracy": 0.8924050632911392, | |
| "eval_loss": 0.6527729630470276, | |
| "eval_runtime": 75.8118, | |
| "eval_samples_per_second": 2.084, | |
| "eval_steps_per_second": 2.084, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "eval_accuracy": 0.8924050632911392, | |
| "eval_loss": 0.6527729630470276, | |
| "eval_runtime": 76.1131, | |
| "eval_samples_per_second": 2.076, | |
| "eval_steps_per_second": 2.076, | |
| "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 | |
| } | |