| { | |
| "best_metric": 0.7872340425531915, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-62", | |
| "epoch": 3.237704918032787, | |
| "eval_steps": 500, | |
| "global_step": 244, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.040983606557377046, | |
| "grad_norm": 5.660297393798828, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3242, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08196721311475409, | |
| "grad_norm": 31.36193084716797, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2215, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12295081967213115, | |
| "grad_norm": 3.619921922683716, | |
| "learning_rate": 4.8858447488584476e-05, | |
| "loss": 0.3642, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16393442622950818, | |
| "grad_norm": 2.1614866256713867, | |
| "learning_rate": 4.657534246575342e-05, | |
| "loss": 1.1585, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.20491803278688525, | |
| "grad_norm": 5.980413436889648, | |
| "learning_rate": 4.4292237442922375e-05, | |
| "loss": 0.5672, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2459016393442623, | |
| "grad_norm": 20.064746856689453, | |
| "learning_rate": 4.200913242009132e-05, | |
| "loss": 0.4643, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2540983606557377, | |
| "eval_accuracy": 0.7872340425531915, | |
| "eval_loss": 0.7600494027137756, | |
| "eval_runtime": 3.3948, | |
| "eval_samples_per_second": 13.845, | |
| "eval_steps_per_second": 3.535, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.0327868852459017, | |
| "grad_norm": 32.691917419433594, | |
| "learning_rate": 3.9726027397260274e-05, | |
| "loss": 0.6524, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.0737704918032787, | |
| "grad_norm": 27.927385330200195, | |
| "learning_rate": 3.744292237442922e-05, | |
| "loss": 0.4143, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.1147540983606556, | |
| "grad_norm": 73.20063781738281, | |
| "learning_rate": 3.5159817351598174e-05, | |
| "loss": 0.5154, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1557377049180328, | |
| "grad_norm": 14.900442123413086, | |
| "learning_rate": 3.287671232876712e-05, | |
| "loss": 0.7259, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1967213114754098, | |
| "grad_norm": 15.045758247375488, | |
| "learning_rate": 3.059360730593607e-05, | |
| "loss": 0.2435, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2377049180327868, | |
| "grad_norm": 1.9350297451019287, | |
| "learning_rate": 2.8310502283105023e-05, | |
| "loss": 0.479, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.2540983606557377, | |
| "eval_accuracy": 0.5957446808510638, | |
| "eval_loss": 1.3025503158569336, | |
| "eval_runtime": 3.3706, | |
| "eval_samples_per_second": 13.944, | |
| "eval_steps_per_second": 3.56, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 2.0245901639344264, | |
| "grad_norm": 80.35542297363281, | |
| "learning_rate": 2.6027397260273973e-05, | |
| "loss": 0.4446, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.0655737704918034, | |
| "grad_norm": 48.855430603027344, | |
| "learning_rate": 2.3744292237442922e-05, | |
| "loss": 0.3023, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.1065573770491803, | |
| "grad_norm": 5.501039028167725, | |
| "learning_rate": 2.1461187214611872e-05, | |
| "loss": 0.0495, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.1475409836065573, | |
| "grad_norm": 42.64323425292969, | |
| "learning_rate": 1.9178082191780822e-05, | |
| "loss": 0.2239, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.1885245901639343, | |
| "grad_norm": 0.1286790668964386, | |
| "learning_rate": 1.689497716894977e-05, | |
| "loss": 0.2566, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.2295081967213113, | |
| "grad_norm": 0.8700327277183533, | |
| "learning_rate": 1.4611872146118721e-05, | |
| "loss": 0.3048, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.2540983606557377, | |
| "eval_accuracy": 0.6808510638297872, | |
| "eval_loss": 0.9859520196914673, | |
| "eval_runtime": 3.3691, | |
| "eval_samples_per_second": 13.95, | |
| "eval_steps_per_second": 3.562, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 3.0163934426229506, | |
| "grad_norm": 10.654744148254395, | |
| "learning_rate": 1.2328767123287671e-05, | |
| "loss": 0.1189, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.057377049180328, | |
| "grad_norm": 0.2807650864124298, | |
| "learning_rate": 1.004566210045662e-05, | |
| "loss": 0.1767, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.098360655737705, | |
| "grad_norm": 29.071613311767578, | |
| "learning_rate": 7.76255707762557e-06, | |
| "loss": 0.233, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.139344262295082, | |
| "grad_norm": 7.512803554534912, | |
| "learning_rate": 5.479452054794521e-06, | |
| "loss": 0.0168, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.180327868852459, | |
| "grad_norm": 1.4460692405700684, | |
| "learning_rate": 3.19634703196347e-06, | |
| "loss": 0.1046, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.221311475409836, | |
| "grad_norm": 0.054721295833587646, | |
| "learning_rate": 9.132420091324201e-07, | |
| "loss": 0.2324, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.237704918032787, | |
| "eval_accuracy": 0.7659574468085106, | |
| "eval_loss": 0.98479163646698, | |
| "eval_runtime": 3.9531, | |
| "eval_samples_per_second": 11.89, | |
| "eval_steps_per_second": 3.036, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.237704918032787, | |
| "step": 244, | |
| "total_flos": 1.2049981873389896e+18, | |
| "train_loss": 0.3545878718866677, | |
| "train_runtime": 283.1012, | |
| "train_samples_per_second": 3.448, | |
| "train_steps_per_second": 0.862 | |
| }, | |
| { | |
| "epoch": 3.237704918032787, | |
| "eval_accuracy": 0.7058823529411765, | |
| "eval_loss": 0.8797193765640259, | |
| "eval_runtime": 7.4733, | |
| "eval_samples_per_second": 13.649, | |
| "eval_steps_per_second": 3.479, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.237704918032787, | |
| "eval_accuracy": 0.7058823529411765, | |
| "eval_loss": 0.8797194361686707, | |
| "eval_runtime": 7.5173, | |
| "eval_samples_per_second": 13.569, | |
| "eval_steps_per_second": 3.459, | |
| "step": 244 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 244, | |
| "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": 1.2049981873389896e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |