| { | |
| "best_metric": 0.8529411764705882, | |
| "best_model_checkpoint": "deit-tiny-patch16-224-finetuned-papsmear/checkpoint-269", | |
| "epoch": 14.805194805194805, | |
| "eval_steps": 500, | |
| "global_step": 285, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5194805194805194, | |
| "grad_norm": 5.887332439422607, | |
| "learning_rate": 1.7241379310344828e-05, | |
| "loss": 1.8247, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.987012987012987, | |
| "eval_accuracy": 0.3014705882352941, | |
| "eval_loss": 1.6199148893356323, | |
| "eval_runtime": 27.9963, | |
| "eval_samples_per_second": 4.858, | |
| "eval_steps_per_second": 0.321, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.0389610389610389, | |
| "grad_norm": 7.520359039306641, | |
| "learning_rate": 3.4482758620689657e-05, | |
| "loss": 1.6044, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.5584415584415585, | |
| "grad_norm": 8.020193099975586, | |
| "learning_rate": 4.9804687500000004e-05, | |
| "loss": 1.415, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.974025974025974, | |
| "eval_accuracy": 0.5147058823529411, | |
| "eval_loss": 1.2593767642974854, | |
| "eval_runtime": 28.328, | |
| "eval_samples_per_second": 4.801, | |
| "eval_steps_per_second": 0.318, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.0779220779220777, | |
| "grad_norm": 9.726988792419434, | |
| "learning_rate": 4.78515625e-05, | |
| "loss": 1.2695, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.5974025974025974, | |
| "grad_norm": 10.028353691101074, | |
| "learning_rate": 4.58984375e-05, | |
| "loss": 1.06, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.961038961038961, | |
| "eval_accuracy": 0.6470588235294118, | |
| "eval_loss": 1.0316276550292969, | |
| "eval_runtime": 28.6395, | |
| "eval_samples_per_second": 4.749, | |
| "eval_steps_per_second": 0.314, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 3.116883116883117, | |
| "grad_norm": 11.8433837890625, | |
| "learning_rate": 4.3945312500000005e-05, | |
| "loss": 1.0001, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 9.538862228393555, | |
| "learning_rate": 4.1992187500000003e-05, | |
| "loss": 0.8808, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.625, | |
| "eval_loss": 1.0088493824005127, | |
| "eval_runtime": 28.6257, | |
| "eval_samples_per_second": 4.751, | |
| "eval_steps_per_second": 0.314, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 4.1558441558441555, | |
| "grad_norm": 13.59125804901123, | |
| "learning_rate": 4.00390625e-05, | |
| "loss": 0.9073, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.675324675324675, | |
| "grad_norm": 10.73096752166748, | |
| "learning_rate": 3.80859375e-05, | |
| "loss": 0.7646, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.987012987012987, | |
| "eval_accuracy": 0.6985294117647058, | |
| "eval_loss": 0.8210764527320862, | |
| "eval_runtime": 28.1022, | |
| "eval_samples_per_second": 4.839, | |
| "eval_steps_per_second": 0.32, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 5.194805194805195, | |
| "grad_norm": 9.227697372436523, | |
| "learning_rate": 3.6132812500000005e-05, | |
| "loss": 0.7113, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 9.764923095703125, | |
| "learning_rate": 3.41796875e-05, | |
| "loss": 0.6798, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.974025974025974, | |
| "eval_accuracy": 0.7132352941176471, | |
| "eval_loss": 0.7382610440254211, | |
| "eval_runtime": 28.5068, | |
| "eval_samples_per_second": 4.771, | |
| "eval_steps_per_second": 0.316, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 6.233766233766234, | |
| "grad_norm": 6.136936187744141, | |
| "learning_rate": 3.22265625e-05, | |
| "loss": 0.6017, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.753246753246753, | |
| "grad_norm": 14.086496353149414, | |
| "learning_rate": 3.02734375e-05, | |
| "loss": 0.554, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.961038961038961, | |
| "eval_accuracy": 0.7573529411764706, | |
| "eval_loss": 0.6476972103118896, | |
| "eval_runtime": 28.4917, | |
| "eval_samples_per_second": 4.773, | |
| "eval_steps_per_second": 0.316, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 10.119462966918945, | |
| "learning_rate": 2.83203125e-05, | |
| "loss": 0.581, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.792207792207792, | |
| "grad_norm": 9.049328804016113, | |
| "learning_rate": 2.63671875e-05, | |
| "loss": 0.5358, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7647058823529411, | |
| "eval_loss": 0.5824333429336548, | |
| "eval_runtime": 28.7788, | |
| "eval_samples_per_second": 4.726, | |
| "eval_steps_per_second": 0.313, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 8.311688311688311, | |
| "grad_norm": 14.884687423706055, | |
| "learning_rate": 2.44140625e-05, | |
| "loss": 0.4674, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.831168831168831, | |
| "grad_norm": 11.825162887573242, | |
| "learning_rate": 2.24609375e-05, | |
| "loss": 0.4689, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.987012987012987, | |
| "eval_accuracy": 0.7794117647058824, | |
| "eval_loss": 0.5570951104164124, | |
| "eval_runtime": 28.5562, | |
| "eval_samples_per_second": 4.763, | |
| "eval_steps_per_second": 0.315, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 9.35064935064935, | |
| "grad_norm": 9.966752052307129, | |
| "learning_rate": 2.05078125e-05, | |
| "loss": 0.414, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.87012987012987, | |
| "grad_norm": 10.21314525604248, | |
| "learning_rate": 1.85546875e-05, | |
| "loss": 0.4217, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.974025974025974, | |
| "eval_accuracy": 0.7867647058823529, | |
| "eval_loss": 0.5505570769309998, | |
| "eval_runtime": 28.6499, | |
| "eval_samples_per_second": 4.747, | |
| "eval_steps_per_second": 0.314, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 10.38961038961039, | |
| "grad_norm": 8.933082580566406, | |
| "learning_rate": 1.66015625e-05, | |
| "loss": 0.3979, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 12.334419250488281, | |
| "learning_rate": 1.4648437500000001e-05, | |
| "loss": 0.4063, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 10.96103896103896, | |
| "eval_accuracy": 0.8235294117647058, | |
| "eval_loss": 0.4987229108810425, | |
| "eval_runtime": 28.2377, | |
| "eval_samples_per_second": 4.816, | |
| "eval_steps_per_second": 0.319, | |
| "step": 211 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 10.463237762451172, | |
| "learning_rate": 1.2695312500000001e-05, | |
| "loss": 0.3799, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 11.948051948051948, | |
| "grad_norm": 9.576591491699219, | |
| "learning_rate": 1.0742187500000001e-05, | |
| "loss": 0.3827, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8088235294117647, | |
| "eval_loss": 0.47928401827812195, | |
| "eval_runtime": 29.299, | |
| "eval_samples_per_second": 4.642, | |
| "eval_steps_per_second": 0.307, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 12.467532467532468, | |
| "grad_norm": 7.557727813720703, | |
| "learning_rate": 8.789062500000001e-06, | |
| "loss": 0.3419, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.987012987012987, | |
| "grad_norm": 7.865088939666748, | |
| "learning_rate": 6.8359375e-06, | |
| "loss": 0.3095, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.987012987012987, | |
| "eval_accuracy": 0.8014705882352942, | |
| "eval_loss": 0.47241097688674927, | |
| "eval_runtime": 28.3959, | |
| "eval_samples_per_second": 4.789, | |
| "eval_steps_per_second": 0.317, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 13.506493506493506, | |
| "grad_norm": 15.5108060836792, | |
| "learning_rate": 4.8828125e-06, | |
| "loss": 0.3521, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 13.974025974025974, | |
| "eval_accuracy": 0.8529411764705882, | |
| "eval_loss": 0.4389486610889435, | |
| "eval_runtime": 28.274, | |
| "eval_samples_per_second": 4.81, | |
| "eval_steps_per_second": 0.318, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 14.025974025974026, | |
| "grad_norm": 9.100951194763184, | |
| "learning_rate": 2.9296875e-06, | |
| "loss": 0.3252, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 14.545454545454545, | |
| "grad_norm": 11.608354568481445, | |
| "learning_rate": 9.765625e-07, | |
| "loss": 0.3397, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 14.805194805194805, | |
| "eval_accuracy": 0.8455882352941176, | |
| "eval_loss": 0.43828412890434265, | |
| "eval_runtime": 28.4264, | |
| "eval_samples_per_second": 4.784, | |
| "eval_steps_per_second": 0.317, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 14.805194805194805, | |
| "step": 285, | |
| "total_flos": 9.04681758989353e+16, | |
| "train_loss": 0.6850269221422965, | |
| "train_runtime": 4202.2024, | |
| "train_samples_per_second": 4.369, | |
| "train_steps_per_second": 0.068 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 285, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "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": 9.04681758989353e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |