| { | |
| "best_metric": 0.7875, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/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": 12.65776252746582, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 1.7008, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05319148936170213, | |
| "grad_norm": 6.046972274780273, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.9981, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0797872340425532, | |
| "grad_norm": 9.531058311462402, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.9371, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.10638297872340426, | |
| "grad_norm": 6.416304111480713, | |
| "learning_rate": 4.970414201183432e-05, | |
| "loss": 0.8702, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.13297872340425532, | |
| "grad_norm": 4.882875442504883, | |
| "learning_rate": 4.822485207100592e-05, | |
| "loss": 0.5812, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1595744680851064, | |
| "grad_norm": 3.9771392345428467, | |
| "learning_rate": 4.674556213017752e-05, | |
| "loss": 1.1064, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.18617021276595744, | |
| "grad_norm": 2.8228275775909424, | |
| "learning_rate": 4.5266272189349114e-05, | |
| "loss": 0.6824, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2127659574468085, | |
| "grad_norm": 2.11715030670166, | |
| "learning_rate": 4.378698224852072e-05, | |
| "loss": 0.5582, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2393617021276596, | |
| "grad_norm": 1.6437604427337646, | |
| "learning_rate": 4.230769230769231e-05, | |
| "loss": 0.7968, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2526595744680851, | |
| "eval_accuracy": 0.7875, | |
| "eval_loss": 0.7984516620635986, | |
| "eval_runtime": 15.2987, | |
| "eval_samples_per_second": 10.458, | |
| "eval_steps_per_second": 1.307, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.0132978723404256, | |
| "grad_norm": 1.7461698055267334, | |
| "learning_rate": 4.0828402366863904e-05, | |
| "loss": 0.6967, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0398936170212767, | |
| "grad_norm": 2.8989648818969727, | |
| "learning_rate": 3.934911242603551e-05, | |
| "loss": 0.8106, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0664893617021276, | |
| "grad_norm": 2.848541021347046, | |
| "learning_rate": 3.7869822485207104e-05, | |
| "loss": 0.741, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0930851063829787, | |
| "grad_norm": 2.6322543621063232, | |
| "learning_rate": 3.63905325443787e-05, | |
| "loss": 0.7027, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.1196808510638299, | |
| "grad_norm": 7.683568954467773, | |
| "learning_rate": 3.49112426035503e-05, | |
| "loss": 0.6579, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1462765957446808, | |
| "grad_norm": 1.7948863506317139, | |
| "learning_rate": 3.3431952662721895e-05, | |
| "loss": 0.8164, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.172872340425532, | |
| "grad_norm": 1.682706594467163, | |
| "learning_rate": 3.195266272189349e-05, | |
| "loss": 0.6967, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.199468085106383, | |
| "grad_norm": 3.870861530303955, | |
| "learning_rate": 3.047337278106509e-05, | |
| "loss": 0.7218, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.226063829787234, | |
| "grad_norm": 1.6848540306091309, | |
| "learning_rate": 2.8994082840236685e-05, | |
| "loss": 0.7805, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.252659574468085, | |
| "grad_norm": 5.371254920959473, | |
| "learning_rate": 2.751479289940829e-05, | |
| "loss": 0.8744, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.252659574468085, | |
| "eval_accuracy": 0.7875, | |
| "eval_loss": 0.7836956977844238, | |
| "eval_runtime": 11.0004, | |
| "eval_samples_per_second": 14.545, | |
| "eval_steps_per_second": 1.818, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.026595744680851, | |
| "grad_norm": 1.314697027206421, | |
| "learning_rate": 2.6035502958579882e-05, | |
| "loss": 0.7122, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0531914893617023, | |
| "grad_norm": 1.7407877445220947, | |
| "learning_rate": 2.4556213017751482e-05, | |
| "loss": 0.7189, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.0797872340425534, | |
| "grad_norm": 2.266850471496582, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.8913, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.106382978723404, | |
| "grad_norm": 2.1507058143615723, | |
| "learning_rate": 2.1597633136094676e-05, | |
| "loss": 0.8554, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.132978723404255, | |
| "grad_norm": 4.038496971130371, | |
| "learning_rate": 2.0118343195266273e-05, | |
| "loss": 0.6919, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.1595744680851063, | |
| "grad_norm": 4.005491256713867, | |
| "learning_rate": 1.8639053254437873e-05, | |
| "loss": 0.7086, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.1861702127659575, | |
| "grad_norm": 3.721208095550537, | |
| "learning_rate": 1.7159763313609466e-05, | |
| "loss": 0.7478, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.2127659574468086, | |
| "grad_norm": 4.95754337310791, | |
| "learning_rate": 1.5680473372781066e-05, | |
| "loss": 0.5457, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.2393617021276597, | |
| "grad_norm": 2.7324411869049072, | |
| "learning_rate": 1.4201183431952663e-05, | |
| "loss": 0.9479, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.252659574468085, | |
| "eval_accuracy": 0.7875, | |
| "eval_loss": 0.7812246084213257, | |
| "eval_runtime": 11.2394, | |
| "eval_samples_per_second": 14.236, | |
| "eval_steps_per_second": 1.779, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 3.0132978723404253, | |
| "grad_norm": 2.182974338531494, | |
| "learning_rate": 1.2721893491124262e-05, | |
| "loss": 0.6173, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.0398936170212765, | |
| "grad_norm": 3.512352228164673, | |
| "learning_rate": 1.1242603550295859e-05, | |
| "loss": 0.6313, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0664893617021276, | |
| "grad_norm": 2.436189651489258, | |
| "learning_rate": 9.763313609467455e-06, | |
| "loss": 0.6465, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.0930851063829787, | |
| "grad_norm": 4.008547782897949, | |
| "learning_rate": 8.284023668639054e-06, | |
| "loss": 0.8407, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.11968085106383, | |
| "grad_norm": 4.314539909362793, | |
| "learning_rate": 6.8047337278106515e-06, | |
| "loss": 0.7826, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.146276595744681, | |
| "grad_norm": 2.1060965061187744, | |
| "learning_rate": 5.325443786982249e-06, | |
| "loss": 0.8714, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.172872340425532, | |
| "grad_norm": 3.250807046890259, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.6124, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.199468085106383, | |
| "grad_norm": 4.283337593078613, | |
| "learning_rate": 2.366863905325444e-06, | |
| "loss": 0.8395, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.226063829787234, | |
| "grad_norm": 4.142538070678711, | |
| "learning_rate": 8.875739644970415e-07, | |
| "loss": 0.8282, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.242021276595745, | |
| "eval_accuracy": 0.7875, | |
| "eval_loss": 0.7884188294410706, | |
| "eval_runtime": 11.8341, | |
| "eval_samples_per_second": 13.52, | |
| "eval_steps_per_second": 1.69, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 3.242021276595745, | |
| "step": 376, | |
| "total_flos": 3.7220613152994755e+18, | |
| "train_loss": 0.785655234088289, | |
| "train_runtime": 616.1212, | |
| "train_samples_per_second": 4.882, | |
| "train_steps_per_second": 0.61 | |
| }, | |
| { | |
| "epoch": 3.242021276595745, | |
| "eval_accuracy": 0.7633136094674556, | |
| "eval_loss": 0.8817278742790222, | |
| "eval_runtime": 19.9254, | |
| "eval_samples_per_second": 8.482, | |
| "eval_steps_per_second": 1.104, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 3.242021276595745, | |
| "eval_accuracy": 0.7633136094674556, | |
| "eval_loss": 0.8817277550697327, | |
| "eval_runtime": 14.8211, | |
| "eval_samples_per_second": 11.403, | |
| "eval_steps_per_second": 1.484, | |
| "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 | |
| } | |