| { | |
| "best_metric": 0.7115384615384616, | |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-2/checkpoint-36", | |
| "epoch": 40.0, | |
| "eval_steps": 500, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.346920132637024, | |
| "eval_runtime": 1.3119, | |
| "eval_samples_per_second": 39.637, | |
| "eval_steps_per_second": 1.525, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4807692307692308, | |
| "eval_loss": 1.3199608325958252, | |
| "eval_runtime": 0.9192, | |
| "eval_samples_per_second": 56.574, | |
| "eval_steps_per_second": 2.176, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4807692307692308, | |
| "eval_loss": 1.3124021291732788, | |
| "eval_runtime": 1.2398, | |
| "eval_samples_per_second": 41.943, | |
| "eval_steps_per_second": 1.613, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.217806100845337, | |
| "eval_runtime": 0.9102, | |
| "eval_samples_per_second": 57.131, | |
| "eval_steps_per_second": 2.197, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 5.100837707519531, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 1.1551, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5769230769230769, | |
| "eval_loss": 1.095713496208191, | |
| "eval_runtime": 0.9082, | |
| "eval_samples_per_second": 57.254, | |
| "eval_steps_per_second": 2.202, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5769230769230769, | |
| "eval_loss": 1.0359364748001099, | |
| "eval_runtime": 0.9173, | |
| "eval_samples_per_second": 56.687, | |
| "eval_steps_per_second": 2.18, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5961538461538461, | |
| "eval_loss": 1.0102628469467163, | |
| "eval_runtime": 1.2573, | |
| "eval_samples_per_second": 41.358, | |
| "eval_steps_per_second": 1.591, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6538461538461539, | |
| "eval_loss": 0.9382320046424866, | |
| "eval_runtime": 0.9221, | |
| "eval_samples_per_second": 56.39, | |
| "eval_steps_per_second": 2.169, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6346153846153846, | |
| "eval_loss": 0.8748425841331482, | |
| "eval_runtime": 0.9162, | |
| "eval_samples_per_second": 56.758, | |
| "eval_steps_per_second": 2.183, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 14.212567329406738, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.9827, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6153846153846154, | |
| "eval_loss": 0.8836320042610168, | |
| "eval_runtime": 0.8982, | |
| "eval_samples_per_second": 57.895, | |
| "eval_steps_per_second": 2.227, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6153846153846154, | |
| "eval_loss": 0.8573623895645142, | |
| "eval_runtime": 1.1544, | |
| "eval_samples_per_second": 45.044, | |
| "eval_steps_per_second": 1.732, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5961538461538461, | |
| "eval_loss": 0.8493903279304504, | |
| "eval_runtime": 0.9118, | |
| "eval_samples_per_second": 57.032, | |
| "eval_steps_per_second": 2.194, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.6153846153846154, | |
| "eval_loss": 0.8225612044334412, | |
| "eval_runtime": 0.9242, | |
| "eval_samples_per_second": 56.266, | |
| "eval_steps_per_second": 2.164, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6346153846153846, | |
| "eval_loss": 0.824204683303833, | |
| "eval_runtime": 1.0428, | |
| "eval_samples_per_second": 49.868, | |
| "eval_steps_per_second": 1.918, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 10.8810396194458, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.8007, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.6153846153846154, | |
| "eval_loss": 0.8303971886634827, | |
| "eval_runtime": 0.9217, | |
| "eval_samples_per_second": 56.42, | |
| "eval_steps_per_second": 2.17, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.6538461538461539, | |
| "eval_loss": 0.8447372913360596, | |
| "eval_runtime": 0.9276, | |
| "eval_samples_per_second": 56.061, | |
| "eval_steps_per_second": 2.156, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.8228283524513245, | |
| "eval_runtime": 0.9517, | |
| "eval_samples_per_second": 54.638, | |
| "eval_steps_per_second": 2.101, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7115384615384616, | |
| "eval_loss": 0.792805016040802, | |
| "eval_runtime": 0.9507, | |
| "eval_samples_per_second": 54.698, | |
| "eval_steps_per_second": 2.104, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.6730769230769231, | |
| "eval_loss": 0.7822188138961792, | |
| "eval_runtime": 1.2729, | |
| "eval_samples_per_second": 40.853, | |
| "eval_steps_per_second": 1.571, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 8.834497451782227, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.6882, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6538461538461539, | |
| "eval_loss": 0.7749820351600647, | |
| "eval_runtime": 0.92, | |
| "eval_samples_per_second": 56.524, | |
| "eval_steps_per_second": 2.174, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.6538461538461539, | |
| "eval_loss": 0.7726113796234131, | |
| "eval_runtime": 0.9301, | |
| "eval_samples_per_second": 55.907, | |
| "eval_steps_per_second": 2.15, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.6730769230769231, | |
| "eval_loss": 0.7897683382034302, | |
| "eval_runtime": 0.9198, | |
| "eval_samples_per_second": 56.533, | |
| "eval_steps_per_second": 2.174, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.6730769230769231, | |
| "eval_loss": 0.8020899891853333, | |
| "eval_runtime": 1.1676, | |
| "eval_samples_per_second": 44.534, | |
| "eval_steps_per_second": 1.713, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.7834421396255493, | |
| "eval_runtime": 0.952, | |
| "eval_samples_per_second": 54.625, | |
| "eval_steps_per_second": 2.101, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 6.992414951324463, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.6154, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.6730769230769231, | |
| "eval_loss": 0.7633832097053528, | |
| "eval_runtime": 0.9404, | |
| "eval_samples_per_second": 55.295, | |
| "eval_steps_per_second": 2.127, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.7583783864974976, | |
| "eval_runtime": 0.9339, | |
| "eval_samples_per_second": 55.678, | |
| "eval_steps_per_second": 2.141, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.6538461538461539, | |
| "eval_loss": 0.7772932052612305, | |
| "eval_runtime": 0.9753, | |
| "eval_samples_per_second": 53.318, | |
| "eval_steps_per_second": 2.051, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.6538461538461539, | |
| "eval_loss": 0.7829698324203491, | |
| "eval_runtime": 1.1036, | |
| "eval_samples_per_second": 47.12, | |
| "eval_steps_per_second": 1.812, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.6538461538461539, | |
| "eval_loss": 0.7719097137451172, | |
| "eval_runtime": 0.9261, | |
| "eval_samples_per_second": 56.149, | |
| "eval_steps_per_second": 2.16, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 10.087705612182617, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.541, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.6538461538461539, | |
| "eval_loss": 0.7602768540382385, | |
| "eval_runtime": 0.9394, | |
| "eval_samples_per_second": 55.357, | |
| "eval_steps_per_second": 2.129, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.6730769230769231, | |
| "eval_loss": 0.7497422695159912, | |
| "eval_runtime": 0.9375, | |
| "eval_samples_per_second": 55.469, | |
| "eval_steps_per_second": 2.133, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7115384615384616, | |
| "eval_loss": 0.7381149530410767, | |
| "eval_runtime": 0.9116, | |
| "eval_samples_per_second": 57.042, | |
| "eval_steps_per_second": 2.194, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.7274792790412903, | |
| "eval_runtime": 1.2761, | |
| "eval_samples_per_second": 40.748, | |
| "eval_steps_per_second": 1.567, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.7277012467384338, | |
| "eval_runtime": 1.2551, | |
| "eval_samples_per_second": 41.429, | |
| "eval_steps_per_second": 1.593, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 9.500407218933105, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.5163, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.7271146774291992, | |
| "eval_runtime": 0.9407, | |
| "eval_samples_per_second": 55.278, | |
| "eval_steps_per_second": 2.126, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.7273848652839661, | |
| "eval_runtime": 0.9428, | |
| "eval_samples_per_second": 55.153, | |
| "eval_steps_per_second": 2.121, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.7304191589355469, | |
| "eval_runtime": 0.945, | |
| "eval_samples_per_second": 55.024, | |
| "eval_steps_per_second": 2.116, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.7329342365264893, | |
| "eval_runtime": 0.9371, | |
| "eval_samples_per_second": 55.491, | |
| "eval_steps_per_second": 2.134, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.7351483702659607, | |
| "eval_runtime": 1.2586, | |
| "eval_samples_per_second": 41.316, | |
| "eval_steps_per_second": 1.589, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 11.405285835266113, | |
| "learning_rate": 0.0, | |
| "loss": 0.5183, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.7355996370315552, | |
| "eval_runtime": 1.036, | |
| "eval_samples_per_second": 50.192, | |
| "eval_steps_per_second": 1.93, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "step": 80, | |
| "total_flos": 3.0975271558447104e+17, | |
| "train_loss": 0.7271977722644806, | |
| "train_runtime": 417.7948, | |
| "train_samples_per_second": 22.786, | |
| "train_steps_per_second": 0.191 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 80, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "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": 3.0975271558447104e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |