| { | |
| "best_metric": 0.3229925036430359, | |
| "best_model_checkpoint": "./drive/Shareddrives/CS198-Drones/[v4] Training Output/efficientnet-b0_rice-leaf-disease-augmented-v4_fft/checkpoint-1728", | |
| "epoch": 15.0, | |
| "eval_steps": 64, | |
| "global_step": 1920, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.9023804664611816, | |
| "learning_rate": 7.5e-06, | |
| "loss": 2.0658, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.24496644295302014, | |
| "eval_loss": 1.9986859560012817, | |
| "eval_runtime": 5.8944, | |
| "eval_samples_per_second": 50.557, | |
| "eval_steps_per_second": 0.848, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.8835070133209229, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.9318, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4261744966442953, | |
| "eval_loss": 1.7966947555541992, | |
| "eval_runtime": 6.6797, | |
| "eval_samples_per_second": 44.613, | |
| "eval_steps_per_second": 0.749, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 1.9900165796279907, | |
| "learning_rate": 2.25e-05, | |
| "loss": 1.6451, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_accuracy": 0.5604026845637584, | |
| "eval_loss": 1.4542900323867798, | |
| "eval_runtime": 5.8004, | |
| "eval_samples_per_second": 51.375, | |
| "eval_steps_per_second": 0.862, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.0631637573242188, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2851, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6845637583892618, | |
| "eval_loss": 1.0677685737609863, | |
| "eval_runtime": 6.7517, | |
| "eval_samples_per_second": 44.137, | |
| "eval_steps_per_second": 0.741, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 2.0110857486724854, | |
| "learning_rate": 2.9025243640281226e-05, | |
| "loss": 0.9017, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_accuracy": 0.761744966442953, | |
| "eval_loss": 0.8011165261268616, | |
| "eval_runtime": 5.7914, | |
| "eval_samples_per_second": 51.456, | |
| "eval_steps_per_second": 0.863, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.901694416999817, | |
| "learning_rate": 2.6227661222566516e-05, | |
| "loss": 0.6891, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8053691275167785, | |
| "eval_loss": 0.6283528208732605, | |
| "eval_runtime": 5.4903, | |
| "eval_samples_per_second": 54.278, | |
| "eval_steps_per_second": 0.911, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 2.0307912826538086, | |
| "learning_rate": 2.197084758065653e-05, | |
| "loss": 0.5118, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "eval_accuracy": 0.802013422818792, | |
| "eval_loss": 0.6123932600021362, | |
| "eval_runtime": 6.3097, | |
| "eval_samples_per_second": 47.229, | |
| "eval_steps_per_second": 0.792, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.4103055000305176, | |
| "learning_rate": 1.6808050203829845e-05, | |
| "loss": 0.4458, | |
| "step": 512 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8120805369127517, | |
| "eval_loss": 0.6052291393280029, | |
| "eval_runtime": 6.2959, | |
| "eval_samples_per_second": 47.332, | |
| "eval_steps_per_second": 0.794, | |
| "step": 512 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 1.8820128440856934, | |
| "learning_rate": 1.1410265035686632e-05, | |
| "loss": 0.356, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "eval_accuracy": 0.8288590604026845, | |
| "eval_loss": 0.5032140016555786, | |
| "eval_runtime": 5.9496, | |
| "eval_samples_per_second": 50.087, | |
| "eval_steps_per_second": 0.84, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.9764384031295776, | |
| "learning_rate": 6.479028799032664e-06, | |
| "loss": 0.3255, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8523489932885906, | |
| "eval_loss": 0.44292640686035156, | |
| "eval_runtime": 5.3491, | |
| "eval_samples_per_second": 55.71, | |
| "eval_steps_per_second": 0.935, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "grad_norm": 1.9877854585647583, | |
| "learning_rate": 2.6552420115951546e-06, | |
| "loss": 0.294, | |
| "step": 704 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "eval_accuracy": 0.8489932885906041, | |
| "eval_loss": 0.48286402225494385, | |
| "eval_runtime": 5.3534, | |
| "eval_samples_per_second": 55.665, | |
| "eval_steps_per_second": 0.934, | |
| "step": 704 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.8483781814575195, | |
| "learning_rate": 4.3587273860921985e-07, | |
| "loss": 0.2856, | |
| "step": 768 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8389261744966443, | |
| "eval_loss": 0.48628225922584534, | |
| "eval_runtime": 5.4523, | |
| "eval_samples_per_second": 54.656, | |
| "eval_steps_per_second": 0.917, | |
| "step": 768 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "grad_norm": 1.8444066047668457, | |
| "learning_rate": 2.989063311147081e-05, | |
| "loss": 0.289, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "eval_accuracy": 0.8523489932885906, | |
| "eval_loss": 0.4306166470050812, | |
| "eval_runtime": 5.5498, | |
| "eval_samples_per_second": 53.696, | |
| "eval_steps_per_second": 0.901, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.8012785911560059, | |
| "learning_rate": 2.8281840384798147e-05, | |
| "loss": 0.2357, | |
| "step": 896 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8590604026845637, | |
| "eval_loss": 0.44078949093818665, | |
| "eval_runtime": 5.421, | |
| "eval_samples_per_second": 54.972, | |
| "eval_steps_per_second": 0.922, | |
| "step": 896 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 1.9915043115615845, | |
| "learning_rate": 2.494683987361193e-05, | |
| "loss": 0.183, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "eval_accuracy": 0.8691275167785235, | |
| "eval_loss": 0.38100528717041016, | |
| "eval_runtime": 6.1836, | |
| "eval_samples_per_second": 48.192, | |
| "eval_steps_per_second": 0.809, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.6568807363510132, | |
| "learning_rate": 2.031907330563804e-05, | |
| "loss": 0.1446, | |
| "step": 1024 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8624161073825504, | |
| "eval_loss": 0.3679770529270172, | |
| "eval_runtime": 6.3939, | |
| "eval_samples_per_second": 46.607, | |
| "eval_steps_per_second": 0.782, | |
| "step": 1024 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "grad_norm": 1.1750797033309937, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1094, | |
| "step": 1088 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "eval_accuracy": 0.8825503355704698, | |
| "eval_loss": 0.35567012429237366, | |
| "eval_runtime": 5.773, | |
| "eval_samples_per_second": 51.62, | |
| "eval_steps_per_second": 0.866, | |
| "step": 1088 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.8476396799087524, | |
| "learning_rate": 9.680926694361966e-06, | |
| "loss": 0.102, | |
| "step": 1152 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8825503355704698, | |
| "eval_loss": 0.3554769456386566, | |
| "eval_runtime": 6.2043, | |
| "eval_samples_per_second": 48.031, | |
| "eval_steps_per_second": 0.806, | |
| "step": 1152 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "grad_norm": 1.0202888250350952, | |
| "learning_rate": 5.053160126388081e-06, | |
| "loss": 0.0752, | |
| "step": 1216 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "eval_accuracy": 0.8825503355704698, | |
| "eval_loss": 0.3513905107975006, | |
| "eval_runtime": 6.2026, | |
| "eval_samples_per_second": 48.044, | |
| "eval_steps_per_second": 0.806, | |
| "step": 1216 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.4681854248046875, | |
| "learning_rate": 1.7181596152018497e-06, | |
| "loss": 0.0828, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8691275167785235, | |
| "eval_loss": 0.4008277654647827, | |
| "eval_runtime": 6.251, | |
| "eval_samples_per_second": 47.673, | |
| "eval_steps_per_second": 0.8, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "grad_norm": 0.9785635471343994, | |
| "learning_rate": 1.0936688852918875e-07, | |
| "loss": 0.0758, | |
| "step": 1344 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "eval_accuracy": 0.8791946308724832, | |
| "eval_loss": 0.35087850689888, | |
| "eval_runtime": 5.9492, | |
| "eval_samples_per_second": 50.091, | |
| "eval_steps_per_second": 0.84, | |
| "step": 1344 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.5723533034324646, | |
| "learning_rate": 2.9564127261390784e-05, | |
| "loss": 0.0699, | |
| "step": 1408 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.889261744966443, | |
| "eval_loss": 0.3736562132835388, | |
| "eval_runtime": 6.3982, | |
| "eval_samples_per_second": 46.575, | |
| "eval_steps_per_second": 0.781, | |
| "step": 1408 | |
| }, | |
| { | |
| "epoch": 11.5, | |
| "grad_norm": 1.2858341932296753, | |
| "learning_rate": 2.7344757988404852e-05, | |
| "loss": 0.059, | |
| "step": 1472 | |
| }, | |
| { | |
| "epoch": 11.5, | |
| "eval_accuracy": 0.8859060402684564, | |
| "eval_loss": 0.34882599115371704, | |
| "eval_runtime": 6.2139, | |
| "eval_samples_per_second": 47.957, | |
| "eval_steps_per_second": 0.805, | |
| "step": 1472 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.8073585033416748, | |
| "learning_rate": 2.352097120096733e-05, | |
| "loss": 0.0504, | |
| "step": 1536 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9026845637583892, | |
| "eval_loss": 0.3544101119041443, | |
| "eval_runtime": 5.7224, | |
| "eval_samples_per_second": 52.076, | |
| "eval_steps_per_second": 0.874, | |
| "step": 1536 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 0.7392011880874634, | |
| "learning_rate": 1.8589734964313355e-05, | |
| "loss": 0.0381, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "eval_accuracy": 0.8926174496644296, | |
| "eval_loss": 0.38668006658554077, | |
| "eval_runtime": 5.6939, | |
| "eval_samples_per_second": 52.337, | |
| "eval_steps_per_second": 0.878, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.3402092158794403, | |
| "learning_rate": 1.319194979617016e-05, | |
| "loss": 0.0293, | |
| "step": 1664 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.889261744966443, | |
| "eval_loss": 0.35926553606987, | |
| "eval_runtime": 5.3061, | |
| "eval_samples_per_second": 56.161, | |
| "eval_steps_per_second": 0.942, | |
| "step": 1664 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "grad_norm": 0.38678640127182007, | |
| "learning_rate": 8.029152419343479e-06, | |
| "loss": 0.0257, | |
| "step": 1728 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "eval_accuracy": 0.8993288590604027, | |
| "eval_loss": 0.3229925036430359, | |
| "eval_runtime": 5.3086, | |
| "eval_samples_per_second": 56.135, | |
| "eval_steps_per_second": 0.942, | |
| "step": 1728 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.9726854562759399, | |
| "learning_rate": 3.772338777433487e-06, | |
| "loss": 0.0258, | |
| "step": 1792 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.889261744966443, | |
| "eval_loss": 0.4092034101486206, | |
| "eval_runtime": 5.3013, | |
| "eval_samples_per_second": 56.212, | |
| "eval_steps_per_second": 0.943, | |
| "step": 1792 | |
| }, | |
| { | |
| "epoch": 14.5, | |
| "grad_norm": 2.582404136657715, | |
| "learning_rate": 9.747563597187792e-07, | |
| "loss": 0.0249, | |
| "step": 1856 | |
| }, | |
| { | |
| "epoch": 14.5, | |
| "eval_accuracy": 0.8859060402684564, | |
| "eval_loss": 0.4290601909160614, | |
| "eval_runtime": 5.279, | |
| "eval_samples_per_second": 56.45, | |
| "eval_steps_per_second": 0.947, | |
| "step": 1856 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.6526421904563904, | |
| "learning_rate": 0.0, | |
| "loss": 0.0187, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8993288590604027, | |
| "eval_loss": 0.3305014669895172, | |
| "eval_runtime": 5.2462, | |
| "eval_samples_per_second": 56.803, | |
| "eval_steps_per_second": 0.953, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 1920, | |
| "total_flos": 4.4590015828721664e+17, | |
| "train_loss": 0.4125567433113853, | |
| "train_runtime": 3265.3981, | |
| "train_samples_per_second": 37.631, | |
| "train_steps_per_second": 0.588 | |
| } | |
| ], | |
| "logging_steps": 64, | |
| "max_steps": 1920, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 64, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.4590015828721664e+17, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |