| { | |
| "best_metric": 1.0, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset\\checkpoint-237", | |
| "epoch": 4.18974358974359, | |
| "eval_steps": 500, | |
| "global_step": 390, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02564102564102564, | |
| "grad_norm": 3.9795846939086914, | |
| "learning_rate": 1.282051282051282e-05, | |
| "loss": 1.2941, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05128205128205128, | |
| "grad_norm": 4.371654510498047, | |
| "learning_rate": 2.564102564102564e-05, | |
| "loss": 0.9964, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07692307692307693, | |
| "grad_norm": 3.8371341228485107, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.6155, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.10256410256410256, | |
| "grad_norm": 6.883916854858398, | |
| "learning_rate": 4.985754985754986e-05, | |
| "loss": 0.2122, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1282051282051282, | |
| "grad_norm": 0.4472762644290924, | |
| "learning_rate": 4.8433048433048433e-05, | |
| "loss": 0.0886, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 1.2847106456756592, | |
| "learning_rate": 4.700854700854701e-05, | |
| "loss": 0.152, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1794871794871795, | |
| "grad_norm": 2.8317058086395264, | |
| "learning_rate": 4.558404558404559e-05, | |
| "loss": 0.1648, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.20256410256410257, | |
| "eval_accuracy": 0.965, | |
| "eval_loss": 0.08287694305181503, | |
| "eval_runtime": 46.4491, | |
| "eval_samples_per_second": 4.306, | |
| "eval_steps_per_second": 0.215, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.0025641025641026, | |
| "grad_norm": 0.3571479618549347, | |
| "learning_rate": 4.415954415954416e-05, | |
| "loss": 0.0616, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.028205128205128, | |
| "grad_norm": 0.1090891882777214, | |
| "learning_rate": 4.2735042735042735e-05, | |
| "loss": 0.0393, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0538461538461539, | |
| "grad_norm": 1.746414065361023, | |
| "learning_rate": 4.131054131054131e-05, | |
| "loss": 0.0735, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0794871794871794, | |
| "grad_norm": 1.1220694780349731, | |
| "learning_rate": 3.988603988603989e-05, | |
| "loss": 0.0842, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.1051282051282052, | |
| "grad_norm": 0.10681042075157166, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.0238, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.1307692307692307, | |
| "grad_norm": 2.7694716453552246, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.089, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.1564102564102563, | |
| "grad_norm": 1.5847387313842773, | |
| "learning_rate": 3.561253561253561e-05, | |
| "loss": 0.019, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.182051282051282, | |
| "grad_norm": 5.06767463684082, | |
| "learning_rate": 3.418803418803419e-05, | |
| "loss": 0.0751, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2025641025641025, | |
| "eval_accuracy": 0.96, | |
| "eval_loss": 0.07137803733348846, | |
| "eval_runtime": 45.3454, | |
| "eval_samples_per_second": 4.411, | |
| "eval_steps_per_second": 0.221, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 2.005128205128205, | |
| "grad_norm": 0.07452063262462616, | |
| "learning_rate": 3.2763532763532764e-05, | |
| "loss": 0.0373, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.0307692307692307, | |
| "grad_norm": 0.04081596061587334, | |
| "learning_rate": 3.133903133903134e-05, | |
| "loss": 0.0629, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.056410256410256, | |
| "grad_norm": 0.3062734305858612, | |
| "learning_rate": 2.9914529914529915e-05, | |
| "loss": 0.0289, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.082051282051282, | |
| "grad_norm": 0.08387012779712677, | |
| "learning_rate": 2.8490028490028492e-05, | |
| "loss": 0.0287, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.1076923076923078, | |
| "grad_norm": 0.26985999941825867, | |
| "learning_rate": 2.706552706552707e-05, | |
| "loss": 0.0052, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.11916792392730713, | |
| "learning_rate": 2.564102564102564e-05, | |
| "loss": 0.0571, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.158974358974359, | |
| "grad_norm": 0.18876294791698456, | |
| "learning_rate": 2.4216524216524217e-05, | |
| "loss": 0.0347, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.184615384615385, | |
| "grad_norm": 0.23059749603271484, | |
| "learning_rate": 2.2792022792022794e-05, | |
| "loss": 0.03, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.2025641025641027, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.007003530394285917, | |
| "eval_runtime": 44.7765, | |
| "eval_samples_per_second": 4.467, | |
| "eval_steps_per_second": 0.223, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 3.0076923076923077, | |
| "grad_norm": 0.06344188004732132, | |
| "learning_rate": 2.1367521367521368e-05, | |
| "loss": 0.005, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.033333333333333, | |
| "grad_norm": 0.05598565936088562, | |
| "learning_rate": 1.9943019943019945e-05, | |
| "loss": 0.0032, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.0589743589743588, | |
| "grad_norm": 0.05869239196181297, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.0211, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.0846153846153848, | |
| "grad_norm": 0.013849779032170773, | |
| "learning_rate": 1.7094017094017095e-05, | |
| "loss": 0.0016, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.1102564102564103, | |
| "grad_norm": 0.019467826932668686, | |
| "learning_rate": 1.566951566951567e-05, | |
| "loss": 0.0018, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.135897435897436, | |
| "grad_norm": 0.012761811725795269, | |
| "learning_rate": 1.4245014245014246e-05, | |
| "loss": 0.0015, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.1615384615384614, | |
| "grad_norm": 0.026674192398786545, | |
| "learning_rate": 1.282051282051282e-05, | |
| "loss": 0.0098, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.1871794871794874, | |
| "grad_norm": 0.06651380658149719, | |
| "learning_rate": 1.1396011396011397e-05, | |
| "loss": 0.0029, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.2025641025641027, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0017288248054683208, | |
| "eval_runtime": 45.3086, | |
| "eval_samples_per_second": 4.414, | |
| "eval_steps_per_second": 0.221, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 4.01025641025641, | |
| "grad_norm": 0.013002458028495312, | |
| "learning_rate": 9.971509971509972e-06, | |
| "loss": 0.002, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.035897435897436, | |
| "grad_norm": 0.014323906041681767, | |
| "learning_rate": 8.547008547008548e-06, | |
| "loss": 0.0014, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.061538461538461, | |
| "grad_norm": 0.05807347595691681, | |
| "learning_rate": 7.122507122507123e-06, | |
| "loss": 0.0014, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.087179487179487, | |
| "grad_norm": 0.013779590837657452, | |
| "learning_rate": 5.6980056980056985e-06, | |
| "loss": 0.0021, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.112820512820512, | |
| "grad_norm": 0.048702094703912735, | |
| "learning_rate": 4.273504273504274e-06, | |
| "loss": 0.0013, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.138461538461539, | |
| "grad_norm": 0.020250266417860985, | |
| "learning_rate": 2.8490028490028492e-06, | |
| "loss": 0.0013, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.164102564102564, | |
| "grad_norm": 0.011229473166167736, | |
| "learning_rate": 1.4245014245014246e-06, | |
| "loss": 0.0012, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.18974358974359, | |
| "grad_norm": 0.014126256108283997, | |
| "learning_rate": 0.0, | |
| "loss": 0.0012, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.18974358974359, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0015335038769990206, | |
| "eval_runtime": 45.755, | |
| "eval_samples_per_second": 4.371, | |
| "eval_steps_per_second": 0.219, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.18974358974359, | |
| "step": 390, | |
| "total_flos": 9.66963368152793e+18, | |
| "train_loss": 0.11110094871180944, | |
| "train_runtime": 2152.4894, | |
| "train_samples_per_second": 3.624, | |
| "train_steps_per_second": 0.181 | |
| }, | |
| { | |
| "epoch": 4.18974358974359, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.005171784199774265, | |
| "eval_runtime": 49.0405, | |
| "eval_samples_per_second": 4.017, | |
| "eval_steps_per_second": 0.204, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.18974358974359, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.005171784199774265, | |
| "eval_runtime": 47.2902, | |
| "eval_samples_per_second": 4.166, | |
| "eval_steps_per_second": 0.211, | |
| "step": 390 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 390, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.66963368152793e+18, | |
| "train_batch_size": 20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |