| { | |
| "best_metric": 0.7875, | |
| "best_model_checkpoint": "videomae-base-finetuned-gesturePhasev1/checkpoint-95", | |
| "epoch": 3.242021276595745, | |
| "eval_steps": 500, | |
| "global_step": 376, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.026595744680851064, | |
| "grad_norm": 3.338482141494751, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.89, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05319148936170213, | |
| "grad_norm": 2.24688458442688, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.9819, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0797872340425532, | |
| "grad_norm": 3.7898941040039062, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.7938, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.10638297872340426, | |
| "grad_norm": 2.4301254749298096, | |
| "learning_rate": 4.970414201183432e-05, | |
| "loss": 0.5596, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.13297872340425532, | |
| "grad_norm": 2.3365397453308105, | |
| "learning_rate": 4.822485207100592e-05, | |
| "loss": 0.5379, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1595744680851064, | |
| "grad_norm": 5.359646320343018, | |
| "learning_rate": 4.674556213017752e-05, | |
| "loss": 0.9426, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.18617021276595744, | |
| "grad_norm": 3.4516568183898926, | |
| "learning_rate": 4.5266272189349114e-05, | |
| "loss": 0.7649, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2127659574468085, | |
| "grad_norm": 7.1903557777404785, | |
| "learning_rate": 4.378698224852072e-05, | |
| "loss": 0.8521, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2393617021276596, | |
| "grad_norm": 3.8120908737182617, | |
| "learning_rate": 4.230769230769231e-05, | |
| "loss": 0.6411, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2526595744680851, | |
| "eval_accuracy": 0.7875, | |
| "eval_loss": 0.7702882885932922, | |
| "eval_runtime": 11.1471, | |
| "eval_samples_per_second": 14.354, | |
| "eval_steps_per_second": 1.794, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.0132978723404256, | |
| "grad_norm": 4.791707992553711, | |
| "learning_rate": 4.0828402366863904e-05, | |
| "loss": 0.6733, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0398936170212767, | |
| "grad_norm": 2.598623514175415, | |
| "learning_rate": 3.934911242603551e-05, | |
| "loss": 0.6375, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0664893617021276, | |
| "grad_norm": 3.0499229431152344, | |
| "learning_rate": 3.7869822485207104e-05, | |
| "loss": 0.8311, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0930851063829787, | |
| "grad_norm": 3.770430088043213, | |
| "learning_rate": 3.63905325443787e-05, | |
| "loss": 0.6421, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.1196808510638299, | |
| "grad_norm": 2.6242294311523438, | |
| "learning_rate": 3.49112426035503e-05, | |
| "loss": 1.0098, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1462765957446808, | |
| "grad_norm": 3.7963359355926514, | |
| "learning_rate": 3.3431952662721895e-05, | |
| "loss": 0.8502, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.172872340425532, | |
| "grad_norm": 5.387338161468506, | |
| "learning_rate": 3.195266272189349e-05, | |
| "loss": 0.7047, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.199468085106383, | |
| "grad_norm": 2.3086750507354736, | |
| "learning_rate": 3.047337278106509e-05, | |
| "loss": 0.6198, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.226063829787234, | |
| "grad_norm": 8.593369483947754, | |
| "learning_rate": 2.8994082840236685e-05, | |
| "loss": 0.7961, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.252659574468085, | |
| "grad_norm": 3.9976279735565186, | |
| "learning_rate": 2.751479289940829e-05, | |
| "loss": 0.7119, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.252659574468085, | |
| "eval_accuracy": 0.7875, | |
| "eval_loss": 0.7828949093818665, | |
| "eval_runtime": 10.79, | |
| "eval_samples_per_second": 14.829, | |
| "eval_steps_per_second": 1.854, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.026595744680851, | |
| "grad_norm": 3.8793869018554688, | |
| "learning_rate": 2.6035502958579882e-05, | |
| "loss": 0.828, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0531914893617023, | |
| "grad_norm": 3.7117443084716797, | |
| "learning_rate": 2.4556213017751482e-05, | |
| "loss": 0.9062, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.0797872340425534, | |
| "grad_norm": 2.0889694690704346, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.6915, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.106382978723404, | |
| "grad_norm": 4.030495643615723, | |
| "learning_rate": 2.1597633136094676e-05, | |
| "loss": 0.6259, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.132978723404255, | |
| "grad_norm": 1.6166176795959473, | |
| "learning_rate": 2.0118343195266273e-05, | |
| "loss": 0.6694, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.1595744680851063, | |
| "grad_norm": 3.049475908279419, | |
| "learning_rate": 1.8639053254437873e-05, | |
| "loss": 0.6296, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.1861702127659575, | |
| "grad_norm": 2.5859758853912354, | |
| "learning_rate": 1.7159763313609466e-05, | |
| "loss": 0.92, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.2127659574468086, | |
| "grad_norm": 3.9732561111450195, | |
| "learning_rate": 1.5680473372781066e-05, | |
| "loss": 0.7367, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.2393617021276597, | |
| "grad_norm": 3.8227388858795166, | |
| "learning_rate": 1.4201183431952663e-05, | |
| "loss": 0.7348, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.252659574468085, | |
| "eval_accuracy": 0.7875, | |
| "eval_loss": 0.8003042936325073, | |
| "eval_runtime": 10.7754, | |
| "eval_samples_per_second": 14.849, | |
| "eval_steps_per_second": 1.856, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 3.0132978723404253, | |
| "grad_norm": 2.1115190982818604, | |
| "learning_rate": 1.2721893491124262e-05, | |
| "loss": 0.7791, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.0398936170212765, | |
| "grad_norm": 4.164553165435791, | |
| "learning_rate": 1.1242603550295859e-05, | |
| "loss": 0.7166, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0664893617021276, | |
| "grad_norm": 3.074124813079834, | |
| "learning_rate": 9.763313609467455e-06, | |
| "loss": 0.6423, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.0930851063829787, | |
| "grad_norm": 7.083850860595703, | |
| "learning_rate": 8.284023668639054e-06, | |
| "loss": 0.7394, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.11968085106383, | |
| "grad_norm": 3.897432804107666, | |
| "learning_rate": 6.8047337278106515e-06, | |
| "loss": 0.7338, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.146276595744681, | |
| "grad_norm": 5.758181571960449, | |
| "learning_rate": 5.325443786982249e-06, | |
| "loss": 0.5162, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.172872340425532, | |
| "grad_norm": 2.1875393390655518, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.7242, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.199468085106383, | |
| "grad_norm": 2.3160412311553955, | |
| "learning_rate": 2.366863905325444e-06, | |
| "loss": 0.6747, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.226063829787234, | |
| "grad_norm": 1.6938382387161255, | |
| "learning_rate": 8.875739644970415e-07, | |
| "loss": 0.6578, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.242021276595745, | |
| "eval_accuracy": 0.7875, | |
| "eval_loss": 0.8131245374679565, | |
| "eval_runtime": 12.412, | |
| "eval_samples_per_second": 12.891, | |
| "eval_steps_per_second": 1.611, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 3.242021276595745, | |
| "step": 376, | |
| "total_flos": 3.7220613152994755e+18, | |
| "train_loss": 0.7423386789382772, | |
| "train_runtime": 620.0465, | |
| "train_samples_per_second": 4.851, | |
| "train_steps_per_second": 0.606 | |
| }, | |
| { | |
| "epoch": 3.242021276595745, | |
| "eval_accuracy": 0.7633136094674556, | |
| "eval_loss": 0.8507111072540283, | |
| "eval_runtime": 13.0467, | |
| "eval_samples_per_second": 12.953, | |
| "eval_steps_per_second": 1.686, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 3.242021276595745, | |
| "eval_accuracy": 0.7633136094674556, | |
| "eval_loss": 0.8507111072540283, | |
| "eval_runtime": 11.1945, | |
| "eval_samples_per_second": 15.097, | |
| "eval_steps_per_second": 1.965, | |
| "step": 376 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 376, | |
| "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": 3.7220613152994755e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |