| { | |
| "best_metric": 0.8714285714285714, | |
| "best_model_checkpoint": "CP2_HAR_vit-base-patch16-224/checkpoint-908", | |
| "epoch": 29.620253164556964, | |
| "eval_steps": 500, | |
| "global_step": 1170, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9873417721518988, | |
| "grad_norm": 2.9084372520446777, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.7032, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.9873417721518988, | |
| "eval_accuracy": 0.3388888888888889, | |
| "eval_loss": 2.304168224334717, | |
| "eval_runtime": 10.2204, | |
| "eval_samples_per_second": 123.283, | |
| "eval_steps_per_second": 1.957, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.045494318008423, | |
| "learning_rate": 3.376068376068376e-05, | |
| "loss": 1.7639, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7515873015873016, | |
| "eval_loss": 1.0595871210098267, | |
| "eval_runtime": 10.2792, | |
| "eval_samples_per_second": 122.577, | |
| "eval_steps_per_second": 1.946, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.9873417721518987, | |
| "grad_norm": 1.7503687143325806, | |
| "learning_rate": 4.995251661918329e-05, | |
| "loss": 0.974, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 2.9873417721518987, | |
| "eval_accuracy": 0.8134920634920635, | |
| "eval_loss": 0.6007124781608582, | |
| "eval_runtime": 10.208, | |
| "eval_samples_per_second": 123.432, | |
| "eval_steps_per_second": 1.959, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.026381492614746, | |
| "learning_rate": 4.8053181386514724e-05, | |
| "loss": 0.7207, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8412698412698413, | |
| "eval_loss": 0.49882617592811584, | |
| "eval_runtime": 10.3437, | |
| "eval_samples_per_second": 121.813, | |
| "eval_steps_per_second": 1.934, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 4.987341772151899, | |
| "grad_norm": 1.8753575086593628, | |
| "learning_rate": 4.620132953466287e-05, | |
| "loss": 0.6285, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 4.987341772151899, | |
| "eval_accuracy": 0.8507936507936508, | |
| "eval_loss": 0.4587480425834656, | |
| "eval_runtime": 10.2618, | |
| "eval_samples_per_second": 122.785, | |
| "eval_steps_per_second": 1.949, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.0627048015594482, | |
| "learning_rate": 4.4301994301994304e-05, | |
| "loss": 0.562, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8531746031746031, | |
| "eval_loss": 0.4662785232067108, | |
| "eval_runtime": 10.2093, | |
| "eval_samples_per_second": 123.417, | |
| "eval_steps_per_second": 1.959, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 6.987341772151899, | |
| "grad_norm": 1.78373122215271, | |
| "learning_rate": 4.2450142450142457e-05, | |
| "loss": 0.5258, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 6.987341772151899, | |
| "eval_accuracy": 0.8468253968253968, | |
| "eval_loss": 0.45184874534606934, | |
| "eval_runtime": 10.2263, | |
| "eval_samples_per_second": 123.211, | |
| "eval_steps_per_second": 1.956, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.4166259765625, | |
| "learning_rate": 4.0550807217473884e-05, | |
| "loss": 0.4843, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8603174603174604, | |
| "eval_loss": 0.4466171860694885, | |
| "eval_runtime": 10.1905, | |
| "eval_samples_per_second": 123.645, | |
| "eval_steps_per_second": 1.963, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 8.987341772151899, | |
| "grad_norm": 2.37298321723938, | |
| "learning_rate": 3.8698955365622036e-05, | |
| "loss": 0.4491, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 8.987341772151899, | |
| "eval_accuracy": 0.8523809523809524, | |
| "eval_loss": 0.43793126940727234, | |
| "eval_runtime": 10.2956, | |
| "eval_samples_per_second": 122.382, | |
| "eval_steps_per_second": 1.943, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.5339529514312744, | |
| "learning_rate": 3.679962013295346e-05, | |
| "loss": 0.4288, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8674603174603175, | |
| "eval_loss": 0.432355135679245, | |
| "eval_runtime": 10.268, | |
| "eval_samples_per_second": 122.712, | |
| "eval_steps_per_second": 1.948, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 10.987341772151899, | |
| "grad_norm": 1.7715898752212524, | |
| "learning_rate": 3.4947768281101616e-05, | |
| "loss": 0.4183, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 10.987341772151899, | |
| "eval_accuracy": 0.8642857142857143, | |
| "eval_loss": 0.44705930352211, | |
| "eval_runtime": 10.2591, | |
| "eval_samples_per_second": 122.817, | |
| "eval_steps_per_second": 1.949, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 2.3940932750701904, | |
| "learning_rate": 3.304843304843305e-05, | |
| "loss": 0.3882, | |
| "step": 474 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8579365079365079, | |
| "eval_loss": 0.43507805466651917, | |
| "eval_runtime": 10.2613, | |
| "eval_samples_per_second": 122.792, | |
| "eval_steps_per_second": 1.949, | |
| "step": 474 | |
| }, | |
| { | |
| "epoch": 12.987341772151899, | |
| "grad_norm": 2.104583501815796, | |
| "learning_rate": 3.1196581196581195e-05, | |
| "loss": 0.3777, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 12.987341772151899, | |
| "eval_accuracy": 0.8611111111111112, | |
| "eval_loss": 0.432034432888031, | |
| "eval_runtime": 10.2065, | |
| "eval_samples_per_second": 123.451, | |
| "eval_steps_per_second": 1.96, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 2.3956658840179443, | |
| "learning_rate": 2.9297245963912633e-05, | |
| "loss": 0.3497, | |
| "step": 553 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8642857142857143, | |
| "eval_loss": 0.4432290196418762, | |
| "eval_runtime": 10.2012, | |
| "eval_samples_per_second": 123.515, | |
| "eval_steps_per_second": 1.961, | |
| "step": 553 | |
| }, | |
| { | |
| "epoch": 14.987341772151899, | |
| "grad_norm": 2.0552070140838623, | |
| "learning_rate": 2.744539411206078e-05, | |
| "loss": 0.347, | |
| "step": 592 | |
| }, | |
| { | |
| "epoch": 14.987341772151899, | |
| "eval_accuracy": 0.8690476190476191, | |
| "eval_loss": 0.4347086548805237, | |
| "eval_runtime": 10.2609, | |
| "eval_samples_per_second": 122.797, | |
| "eval_steps_per_second": 1.949, | |
| "step": 592 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 2.1157126426696777, | |
| "learning_rate": 2.5546058879392216e-05, | |
| "loss": 0.3331, | |
| "step": 632 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8603174603174604, | |
| "eval_loss": 0.4517436623573303, | |
| "eval_runtime": 10.3042, | |
| "eval_samples_per_second": 122.28, | |
| "eval_steps_per_second": 1.941, | |
| "step": 632 | |
| }, | |
| { | |
| "epoch": 16.9873417721519, | |
| "grad_norm": 1.8309712409973145, | |
| "learning_rate": 2.3694207027540365e-05, | |
| "loss": 0.3219, | |
| "step": 671 | |
| }, | |
| { | |
| "epoch": 16.9873417721519, | |
| "eval_accuracy": 0.8666666666666667, | |
| "eval_loss": 0.44011229276657104, | |
| "eval_runtime": 10.2371, | |
| "eval_samples_per_second": 123.082, | |
| "eval_steps_per_second": 1.954, | |
| "step": 671 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 2.178051710128784, | |
| "learning_rate": 2.1794871794871795e-05, | |
| "loss": 0.3081, | |
| "step": 711 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8690476190476191, | |
| "eval_loss": 0.4321274161338806, | |
| "eval_runtime": 10.2691, | |
| "eval_samples_per_second": 122.699, | |
| "eval_steps_per_second": 1.948, | |
| "step": 711 | |
| }, | |
| { | |
| "epoch": 18.9873417721519, | |
| "grad_norm": 2.0867300033569336, | |
| "learning_rate": 1.9943019943019945e-05, | |
| "loss": 0.3194, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 18.9873417721519, | |
| "eval_accuracy": 0.8690476190476191, | |
| "eval_loss": 0.4421131908893585, | |
| "eval_runtime": 10.2636, | |
| "eval_samples_per_second": 122.764, | |
| "eval_steps_per_second": 1.949, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 2.312155246734619, | |
| "learning_rate": 1.804368471035138e-05, | |
| "loss": 0.3102, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8682539682539683, | |
| "eval_loss": 0.4470122754573822, | |
| "eval_runtime": 10.428, | |
| "eval_samples_per_second": 120.829, | |
| "eval_steps_per_second": 1.918, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 20.9873417721519, | |
| "grad_norm": 1.674055814743042, | |
| "learning_rate": 1.6191832858499524e-05, | |
| "loss": 0.2908, | |
| "step": 829 | |
| }, | |
| { | |
| "epoch": 20.9873417721519, | |
| "eval_accuracy": 0.8666666666666667, | |
| "eval_loss": 0.4368663430213928, | |
| "eval_runtime": 10.304, | |
| "eval_samples_per_second": 122.282, | |
| "eval_steps_per_second": 1.941, | |
| "step": 829 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 1.8067846298217773, | |
| "learning_rate": 1.4292497625830961e-05, | |
| "loss": 0.2794, | |
| "step": 869 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8642857142857143, | |
| "eval_loss": 0.4426242411136627, | |
| "eval_runtime": 10.2667, | |
| "eval_samples_per_second": 122.726, | |
| "eval_steps_per_second": 1.948, | |
| "step": 869 | |
| }, | |
| { | |
| "epoch": 22.9873417721519, | |
| "grad_norm": 2.093015193939209, | |
| "learning_rate": 1.2440645773979107e-05, | |
| "loss": 0.2684, | |
| "step": 908 | |
| }, | |
| { | |
| "epoch": 22.9873417721519, | |
| "eval_accuracy": 0.8714285714285714, | |
| "eval_loss": 0.4378375709056854, | |
| "eval_runtime": 10.2839, | |
| "eval_samples_per_second": 122.522, | |
| "eval_steps_per_second": 1.945, | |
| "step": 908 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 1.7382984161376953, | |
| "learning_rate": 1.0541310541310543e-05, | |
| "loss": 0.2635, | |
| "step": 948 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8698412698412699, | |
| "eval_loss": 0.44393062591552734, | |
| "eval_runtime": 10.2502, | |
| "eval_samples_per_second": 122.924, | |
| "eval_steps_per_second": 1.951, | |
| "step": 948 | |
| }, | |
| { | |
| "epoch": 24.9873417721519, | |
| "grad_norm": 1.4845259189605713, | |
| "learning_rate": 8.68945868945869e-06, | |
| "loss": 0.2754, | |
| "step": 987 | |
| }, | |
| { | |
| "epoch": 24.9873417721519, | |
| "eval_accuracy": 0.8642857142857143, | |
| "eval_loss": 0.45485520362854004, | |
| "eval_runtime": 10.2015, | |
| "eval_samples_per_second": 123.511, | |
| "eval_steps_per_second": 1.96, | |
| "step": 987 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 1.6323109865188599, | |
| "learning_rate": 6.790123456790123e-06, | |
| "loss": 0.2669, | |
| "step": 1027 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8674603174603175, | |
| "eval_loss": 0.44393137097358704, | |
| "eval_runtime": 10.2535, | |
| "eval_samples_per_second": 122.885, | |
| "eval_steps_per_second": 1.951, | |
| "step": 1027 | |
| }, | |
| { | |
| "epoch": 26.9873417721519, | |
| "grad_norm": 1.589407205581665, | |
| "learning_rate": 4.938271604938272e-06, | |
| "loss": 0.2616, | |
| "step": 1066 | |
| }, | |
| { | |
| "epoch": 26.9873417721519, | |
| "eval_accuracy": 0.8714285714285714, | |
| "eval_loss": 0.4428676962852478, | |
| "eval_runtime": 10.2429, | |
| "eval_samples_per_second": 123.012, | |
| "eval_steps_per_second": 1.953, | |
| "step": 1066 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 1.8043303489685059, | |
| "learning_rate": 3.038936372269706e-06, | |
| "loss": 0.2501, | |
| "step": 1106 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8698412698412699, | |
| "eval_loss": 0.4408431649208069, | |
| "eval_runtime": 10.1648, | |
| "eval_samples_per_second": 123.957, | |
| "eval_steps_per_second": 1.968, | |
| "step": 1106 | |
| }, | |
| { | |
| "epoch": 28.9873417721519, | |
| "grad_norm": 2.025970935821533, | |
| "learning_rate": 1.1870845204178538e-06, | |
| "loss": 0.2622, | |
| "step": 1145 | |
| }, | |
| { | |
| "epoch": 28.9873417721519, | |
| "eval_accuracy": 0.8682539682539683, | |
| "eval_loss": 0.4434352219104767, | |
| "eval_runtime": 10.25, | |
| "eval_samples_per_second": 122.927, | |
| "eval_steps_per_second": 1.951, | |
| "step": 1145 | |
| }, | |
| { | |
| "epoch": 29.620253164556964, | |
| "grad_norm": 1.5968279838562012, | |
| "learning_rate": 0.0, | |
| "loss": 0.2511, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 29.620253164556964, | |
| "eval_accuracy": 0.8682539682539683, | |
| "eval_loss": 0.44374439120292664, | |
| "eval_runtime": 10.1581, | |
| "eval_samples_per_second": 124.038, | |
| "eval_steps_per_second": 1.969, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 29.620253164556964, | |
| "step": 1170, | |
| "total_flos": 2.3141184141358596e+19, | |
| "train_loss": 0.5155148339067769, | |
| "train_runtime": 6443.0087, | |
| "train_samples_per_second": 46.935, | |
| "train_steps_per_second": 0.182 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1170, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "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": 2.3141184141358596e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |