| { | |
| "best_metric": 0.6666666666666666, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-hasta-55-fold4/checkpoint-35", | |
| "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": 0.4166666666666667, | |
| "eval_loss": 1.1562626361846924, | |
| "eval_runtime": 0.5049, | |
| "eval_samples_per_second": 71.298, | |
| "eval_steps_per_second": 3.961, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "eval_accuracy": 0.3611111111111111, | |
| "eval_loss": 1.11666738986969, | |
| "eval_runtime": 0.5001, | |
| "eval_samples_per_second": 71.99, | |
| "eval_steps_per_second": 3.999, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.3611111111111111, | |
| "eval_loss": 1.3250519037246704, | |
| "eval_runtime": 0.5136, | |
| "eval_samples_per_second": 70.092, | |
| "eval_steps_per_second": 3.894, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3611111111111111, | |
| "eval_loss": 1.140106201171875, | |
| "eval_runtime": 0.5123, | |
| "eval_samples_per_second": 70.265, | |
| "eval_steps_per_second": 3.904, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0346479415893555, | |
| "eval_runtime": 0.5139, | |
| "eval_samples_per_second": 70.055, | |
| "eval_steps_per_second": 3.892, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 4.996661186218262, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1327, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.037699818611145, | |
| "eval_runtime": 0.4965, | |
| "eval_samples_per_second": 72.502, | |
| "eval_steps_per_second": 4.028, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0116735696792603, | |
| "eval_runtime": 0.4907, | |
| "eval_samples_per_second": 73.361, | |
| "eval_steps_per_second": 4.076, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.0720973014831543, | |
| "eval_runtime": 0.4882, | |
| "eval_samples_per_second": 73.747, | |
| "eval_steps_per_second": 4.097, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.0312387943267822, | |
| "eval_runtime": 0.4883, | |
| "eval_samples_per_second": 73.729, | |
| "eval_steps_per_second": 4.096, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0228009223937988, | |
| "eval_runtime": 0.5171, | |
| "eval_samples_per_second": 69.624, | |
| "eval_steps_per_second": 3.868, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9868431091308594, | |
| "eval_runtime": 0.4986, | |
| "eval_samples_per_second": 72.204, | |
| "eval_steps_per_second": 4.011, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 4.689541339874268, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.9623, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.1119240522384644, | |
| "eval_runtime": 0.5133, | |
| "eval_samples_per_second": 70.138, | |
| "eval_steps_per_second": 3.897, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.1094300746917725, | |
| "eval_runtime": 0.4995, | |
| "eval_samples_per_second": 72.068, | |
| "eval_steps_per_second": 4.004, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9811188578605652, | |
| "eval_runtime": 0.493, | |
| "eval_samples_per_second": 73.019, | |
| "eval_steps_per_second": 4.057, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.0887576341629028, | |
| "eval_runtime": 0.4969, | |
| "eval_samples_per_second": 72.45, | |
| "eval_steps_per_second": 4.025, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0576119422912598, | |
| "eval_runtime": 0.4917, | |
| "eval_samples_per_second": 73.217, | |
| "eval_steps_per_second": 4.068, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 16.571428571428573, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 1.058294415473938, | |
| "eval_runtime": 0.4993, | |
| "eval_samples_per_second": 72.098, | |
| "eval_steps_per_second": 4.005, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 5.023176670074463, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.8158, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 17.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0731375217437744, | |
| "eval_runtime": 0.5047, | |
| "eval_samples_per_second": 71.326, | |
| "eval_steps_per_second": 3.963, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9775726795196533, | |
| "eval_runtime": 0.4976, | |
| "eval_samples_per_second": 72.346, | |
| "eval_steps_per_second": 4.019, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 1.0211443901062012, | |
| "eval_runtime": 0.4947, | |
| "eval_samples_per_second": 72.769, | |
| "eval_steps_per_second": 4.043, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 20.571428571428573, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.0035310983657837, | |
| "eval_runtime": 0.4972, | |
| "eval_samples_per_second": 72.403, | |
| "eval_steps_per_second": 4.022, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 21.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9844934940338135, | |
| "eval_runtime": 0.4978, | |
| "eval_samples_per_second": 72.312, | |
| "eval_steps_per_second": 4.017, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 4.650636196136475, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.653, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 1.085641860961914, | |
| "eval_runtime": 0.5034, | |
| "eval_samples_per_second": 71.519, | |
| "eval_steps_per_second": 3.973, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.0787067413330078, | |
| "eval_runtime": 0.5195, | |
| "eval_samples_per_second": 69.3, | |
| "eval_steps_per_second": 3.85, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 24.571428571428573, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.053583025932312, | |
| "eval_runtime": 0.5122, | |
| "eval_samples_per_second": 70.279, | |
| "eval_steps_per_second": 3.904, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0995006561279297, | |
| "eval_runtime": 0.5057, | |
| "eval_samples_per_second": 71.191, | |
| "eval_steps_per_second": 3.955, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.120709776878357, | |
| "eval_runtime": 0.4988, | |
| "eval_samples_per_second": 72.174, | |
| "eval_steps_per_second": 4.01, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.1610081195831299, | |
| "eval_runtime": 0.5118, | |
| "eval_samples_per_second": 70.336, | |
| "eval_steps_per_second": 3.908, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 4.384435653686523, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.5371, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.1163524389266968, | |
| "eval_runtime": 0.5015, | |
| "eval_samples_per_second": 71.781, | |
| "eval_steps_per_second": 3.988, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 29.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.1101287603378296, | |
| "eval_runtime": 0.494, | |
| "eval_samples_per_second": 72.867, | |
| "eval_steps_per_second": 4.048, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.1039893627166748, | |
| "eval_runtime": 0.5228, | |
| "eval_samples_per_second": 68.864, | |
| "eval_steps_per_second": 3.826, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0995060205459595, | |
| "eval_runtime": 0.5075, | |
| "eval_samples_per_second": 70.942, | |
| "eval_steps_per_second": 3.941, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 32.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.1076290607452393, | |
| "eval_runtime": 0.4983, | |
| "eval_samples_per_second": 72.239, | |
| "eval_steps_per_second": 4.013, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 33.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.1530293226242065, | |
| "eval_runtime": 0.5079, | |
| "eval_samples_per_second": 70.874, | |
| "eval_steps_per_second": 3.937, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 4.767789840698242, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.4039, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.177981972694397, | |
| "eval_runtime": 0.4945, | |
| "eval_samples_per_second": 72.803, | |
| "eval_steps_per_second": 4.045, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.13791024684906, | |
| "eval_runtime": 0.4971, | |
| "eval_samples_per_second": 72.427, | |
| "eval_steps_per_second": 4.024, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 36.57142857142857, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.1323542594909668, | |
| "eval_runtime": 0.4999, | |
| "eval_samples_per_second": 72.019, | |
| "eval_steps_per_second": 4.001, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 37.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.154639720916748, | |
| "eval_runtime": 0.5014, | |
| "eval_samples_per_second": 71.793, | |
| "eval_steps_per_second": 3.989, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.1910489797592163, | |
| "eval_runtime": 0.5143, | |
| "eval_samples_per_second": 69.992, | |
| "eval_steps_per_second": 3.888, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 5.7556023597717285, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.348, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.1487263441085815, | |
| "eval_runtime": 0.5177, | |
| "eval_samples_per_second": 69.534, | |
| "eval_steps_per_second": 3.863, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.1367430686950684, | |
| "eval_runtime": 0.4983, | |
| "eval_samples_per_second": 72.242, | |
| "eval_steps_per_second": 4.013, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 41.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.1627492904663086, | |
| "eval_runtime": 0.4964, | |
| "eval_samples_per_second": 72.527, | |
| "eval_steps_per_second": 4.029, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.2421307563781738, | |
| "eval_runtime": 0.5033, | |
| "eval_samples_per_second": 71.534, | |
| "eval_steps_per_second": 3.974, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.307342290878296, | |
| "eval_runtime": 0.509, | |
| "eval_samples_per_second": 70.73, | |
| "eval_steps_per_second": 3.929, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 44.57142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.2772834300994873, | |
| "eval_runtime": 0.498, | |
| "eval_samples_per_second": 72.287, | |
| "eval_steps_per_second": 4.016, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 3.38767409324646, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.321, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.2160472869873047, | |
| "eval_runtime": 0.4997, | |
| "eval_samples_per_second": 72.036, | |
| "eval_steps_per_second": 4.002, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.222046971321106, | |
| "eval_runtime": 0.4972, | |
| "eval_samples_per_second": 72.399, | |
| "eval_steps_per_second": 4.022, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.2167954444885254, | |
| "eval_runtime": 0.5049, | |
| "eval_samples_per_second": 71.297, | |
| "eval_steps_per_second": 3.961, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.218950867652893, | |
| "eval_runtime": 0.5052, | |
| "eval_samples_per_second": 71.264, | |
| "eval_steps_per_second": 3.959, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 49.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.2254923582077026, | |
| "eval_runtime": 0.5043, | |
| "eval_samples_per_second": 71.381, | |
| "eval_steps_per_second": 3.966, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.2173601388931274, | |
| "eval_runtime": 0.4963, | |
| "eval_samples_per_second": 72.541, | |
| "eval_steps_per_second": 4.03, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 3.897371768951416, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.3168, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.2151846885681152, | |
| "eval_runtime": 0.4965, | |
| "eval_samples_per_second": 72.506, | |
| "eval_steps_per_second": 4.028, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 52.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.2308071851730347, | |
| "eval_runtime": 0.5215, | |
| "eval_samples_per_second": 69.031, | |
| "eval_steps_per_second": 3.835, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 53.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.2727937698364258, | |
| "eval_runtime": 0.4971, | |
| "eval_samples_per_second": 72.418, | |
| "eval_steps_per_second": 4.023, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.312135934829712, | |
| "eval_runtime": 0.5009, | |
| "eval_samples_per_second": 71.871, | |
| "eval_steps_per_second": 3.993, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.3349864482879639, | |
| "eval_runtime": 0.4965, | |
| "eval_samples_per_second": 72.504, | |
| "eval_steps_per_second": 4.028, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 56.57142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.3353984355926514, | |
| "eval_runtime": 0.4955, | |
| "eval_samples_per_second": 72.649, | |
| "eval_steps_per_second": 4.036, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 3.2906365394592285, | |
| "learning_rate": 0.0, | |
| "loss": 0.2876, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.3345967531204224, | |
| "eval_runtime": 0.5031, | |
| "eval_samples_per_second": 71.558, | |
| "eval_steps_per_second": 3.975, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "step": 100, | |
| "total_flos": 8.770779002957046e+17, | |
| "train_loss": 0.5778076982498169, | |
| "train_runtime": 552.2846, | |
| "train_samples_per_second": 35.851, | |
| "train_steps_per_second": 0.181 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 1.0211443901062012, | |
| "eval_runtime": 0.5398, | |
| "eval_samples_per_second": 66.687, | |
| "eval_steps_per_second": 3.705, | |
| "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 | |
| } | |