| { | |
| "best_metric": 0.6388888888888888, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-hasta-55-fold5/checkpoint-63", | |
| "epoch": 57.142857142857146, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5714285714285714, | |
| "eval_accuracy": 0.2777777777777778, | |
| "eval_loss": 1.1479322910308838, | |
| "eval_runtime": 0.5044, | |
| "eval_samples_per_second": 71.379, | |
| "eval_steps_per_second": 3.965, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.1506075859069824, | |
| "eval_runtime": 0.4962, | |
| "eval_samples_per_second": 72.554, | |
| "eval_steps_per_second": 4.031, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.212721347808838, | |
| "eval_runtime": 0.5039, | |
| "eval_samples_per_second": 71.439, | |
| "eval_steps_per_second": 3.969, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.051730990409851, | |
| "eval_runtime": 0.5151, | |
| "eval_samples_per_second": 69.891, | |
| "eval_steps_per_second": 3.883, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0113143920898438, | |
| "eval_runtime": 0.503, | |
| "eval_samples_per_second": 71.569, | |
| "eval_steps_per_second": 3.976, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 6.770098686218262, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0752, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.0014044046401978, | |
| "eval_runtime": 0.4985, | |
| "eval_samples_per_second": 72.221, | |
| "eval_steps_per_second": 4.012, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9754472374916077, | |
| "eval_runtime": 0.4887, | |
| "eval_samples_per_second": 73.662, | |
| "eval_steps_per_second": 4.092, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9683561325073242, | |
| "eval_runtime": 0.5007, | |
| "eval_samples_per_second": 71.899, | |
| "eval_steps_per_second": 3.994, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.9598144888877869, | |
| "eval_runtime": 0.4868, | |
| "eval_samples_per_second": 73.958, | |
| "eval_steps_per_second": 4.109, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9349936246871948, | |
| "eval_runtime": 0.4937, | |
| "eval_samples_per_second": 72.924, | |
| "eval_steps_per_second": 4.051, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0847840309143066, | |
| "eval_runtime": 0.5066, | |
| "eval_samples_per_second": 71.068, | |
| "eval_steps_per_second": 3.948, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 6.522934913635254, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.8688, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.9478690028190613, | |
| "eval_runtime": 0.4956, | |
| "eval_samples_per_second": 72.642, | |
| "eval_steps_per_second": 4.036, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 0.9567791819572449, | |
| "eval_runtime": 0.4926, | |
| "eval_samples_per_second": 73.086, | |
| "eval_steps_per_second": 4.06, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.985884964466095, | |
| "eval_runtime": 0.4899, | |
| "eval_samples_per_second": 73.483, | |
| "eval_steps_per_second": 4.082, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0115987062454224, | |
| "eval_runtime": 0.5021, | |
| "eval_samples_per_second": 71.7, | |
| "eval_steps_per_second": 3.983, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.0541839599609375, | |
| "eval_runtime": 0.5039, | |
| "eval_samples_per_second": 71.448, | |
| "eval_steps_per_second": 3.969, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 16.571428571428573, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.9992576241493225, | |
| "eval_runtime": 0.5007, | |
| "eval_samples_per_second": 71.896, | |
| "eval_steps_per_second": 3.994, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 6.972790241241455, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.7636, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 17.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9895581007003784, | |
| "eval_runtime": 0.4882, | |
| "eval_samples_per_second": 73.742, | |
| "eval_steps_per_second": 4.097, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9692603349685669, | |
| "eval_runtime": 0.4901, | |
| "eval_samples_per_second": 73.447, | |
| "eval_steps_per_second": 4.08, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9705440998077393, | |
| "eval_runtime": 0.5325, | |
| "eval_samples_per_second": 67.605, | |
| "eval_steps_per_second": 3.756, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 20.571428571428573, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.999030351638794, | |
| "eval_runtime": 0.5019, | |
| "eval_samples_per_second": 71.722, | |
| "eval_steps_per_second": 3.985, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 21.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0442277193069458, | |
| "eval_runtime": 0.5006, | |
| "eval_samples_per_second": 71.918, | |
| "eval_steps_per_second": 3.995, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 3.7947804927825928, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.5868, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0241245031356812, | |
| "eval_runtime": 0.4938, | |
| "eval_samples_per_second": 72.899, | |
| "eval_steps_per_second": 4.05, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.9458177089691162, | |
| "eval_runtime": 0.497, | |
| "eval_samples_per_second": 72.437, | |
| "eval_steps_per_second": 4.024, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 24.571428571428573, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9390719532966614, | |
| "eval_runtime": 0.4919, | |
| "eval_samples_per_second": 73.193, | |
| "eval_steps_per_second": 4.066, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9626221060752869, | |
| "eval_runtime": 0.5016, | |
| "eval_samples_per_second": 71.773, | |
| "eval_steps_per_second": 3.987, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.969885528087616, | |
| "eval_runtime": 0.4974, | |
| "eval_samples_per_second": 72.37, | |
| "eval_steps_per_second": 4.021, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.9504163265228271, | |
| "eval_runtime": 0.5112, | |
| "eval_samples_per_second": 70.425, | |
| "eval_steps_per_second": 3.912, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 4.607025623321533, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.4816, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 0.9226244688034058, | |
| "eval_runtime": 0.4945, | |
| "eval_samples_per_second": 72.807, | |
| "eval_steps_per_second": 4.045, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 29.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9353493452072144, | |
| "eval_runtime": 0.5076, | |
| "eval_samples_per_second": 70.916, | |
| "eval_steps_per_second": 3.94, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9029681086540222, | |
| "eval_runtime": 0.4988, | |
| "eval_samples_per_second": 72.175, | |
| "eval_steps_per_second": 4.01, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.8947631120681763, | |
| "eval_runtime": 0.4975, | |
| "eval_samples_per_second": 72.361, | |
| "eval_steps_per_second": 4.02, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 32.57142857142857, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9271517992019653, | |
| "eval_runtime": 0.5001, | |
| "eval_samples_per_second": 71.987, | |
| "eval_steps_per_second": 3.999, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 33.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.920202374458313, | |
| "eval_runtime": 0.493, | |
| "eval_samples_per_second": 73.022, | |
| "eval_steps_per_second": 4.057, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 4.770364761352539, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.3909, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9051810503005981, | |
| "eval_runtime": 0.496, | |
| "eval_samples_per_second": 72.581, | |
| "eval_steps_per_second": 4.032, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9257633686065674, | |
| "eval_runtime": 0.4936, | |
| "eval_samples_per_second": 72.933, | |
| "eval_steps_per_second": 4.052, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 36.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.926737904548645, | |
| "eval_runtime": 0.503, | |
| "eval_samples_per_second": 71.575, | |
| "eval_steps_per_second": 3.976, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 37.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9902366995811462, | |
| "eval_runtime": 0.5087, | |
| "eval_samples_per_second": 70.772, | |
| "eval_steps_per_second": 3.932, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0495057106018066, | |
| "eval_runtime": 0.4972, | |
| "eval_samples_per_second": 72.4, | |
| "eval_steps_per_second": 4.022, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 4.217521667480469, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.3124, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9900334477424622, | |
| "eval_runtime": 0.491, | |
| "eval_samples_per_second": 73.314, | |
| "eval_steps_per_second": 4.073, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.57142857142857, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9510228037834167, | |
| "eval_runtime": 0.5072, | |
| "eval_samples_per_second": 70.976, | |
| "eval_steps_per_second": 3.943, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 41.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9530906677246094, | |
| "eval_runtime": 0.5018, | |
| "eval_samples_per_second": 71.744, | |
| "eval_steps_per_second": 3.986, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9438815712928772, | |
| "eval_runtime": 0.5119, | |
| "eval_samples_per_second": 70.325, | |
| "eval_steps_per_second": 3.907, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9520604014396667, | |
| "eval_runtime": 0.4987, | |
| "eval_samples_per_second": 72.186, | |
| "eval_steps_per_second": 4.01, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 44.57142857142857, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9530801773071289, | |
| "eval_runtime": 0.5289, | |
| "eval_samples_per_second": 68.066, | |
| "eval_steps_per_second": 3.781, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 3.9363913536071777, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.3225, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.9550597071647644, | |
| "eval_runtime": 0.494, | |
| "eval_samples_per_second": 72.878, | |
| "eval_steps_per_second": 4.049, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.952035665512085, | |
| "eval_runtime": 0.5179, | |
| "eval_samples_per_second": 69.515, | |
| "eval_steps_per_second": 3.862, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9464225769042969, | |
| "eval_runtime": 0.5093, | |
| "eval_samples_per_second": 70.691, | |
| "eval_steps_per_second": 3.927, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9409239292144775, | |
| "eval_runtime": 0.5199, | |
| "eval_samples_per_second": 69.248, | |
| "eval_steps_per_second": 3.847, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 49.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.947088897228241, | |
| "eval_runtime": 0.5075, | |
| "eval_samples_per_second": 70.93, | |
| "eval_steps_per_second": 3.941, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9645647406578064, | |
| "eval_runtime": 0.5008, | |
| "eval_samples_per_second": 71.887, | |
| "eval_steps_per_second": 3.994, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 5.873734474182129, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.2829, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9804750084877014, | |
| "eval_runtime": 0.498, | |
| "eval_samples_per_second": 72.29, | |
| "eval_steps_per_second": 4.016, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 52.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9747391939163208, | |
| "eval_runtime": 0.5008, | |
| "eval_samples_per_second": 71.881, | |
| "eval_steps_per_second": 3.993, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 53.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9645588397979736, | |
| "eval_runtime": 0.5146, | |
| "eval_samples_per_second": 69.958, | |
| "eval_steps_per_second": 3.887, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.965861439704895, | |
| "eval_runtime": 0.498, | |
| "eval_samples_per_second": 72.29, | |
| "eval_steps_per_second": 4.016, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9643504619598389, | |
| "eval_runtime": 0.5496, | |
| "eval_samples_per_second": 65.5, | |
| "eval_steps_per_second": 3.639, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 56.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9645801186561584, | |
| "eval_runtime": 0.4975, | |
| "eval_samples_per_second": 72.368, | |
| "eval_steps_per_second": 4.02, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 4.704319953918457, | |
| "learning_rate": 0.0, | |
| "loss": 0.272, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.964775800704956, | |
| "eval_runtime": 0.4983, | |
| "eval_samples_per_second": 72.243, | |
| "eval_steps_per_second": 4.014, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "step": 100, | |
| "total_flos": 8.770779002957046e+17, | |
| "train_loss": 0.5356709051132202, | |
| "train_runtime": 558.1878, | |
| "train_samples_per_second": 35.472, | |
| "train_steps_per_second": 0.179 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9257633686065674, | |
| "eval_runtime": 0.5437, | |
| "eval_samples_per_second": 66.216, | |
| "eval_steps_per_second": 3.679, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "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": 8.770779002957046e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |