| { | |
| "best_metric": 0.75, | |
| "best_model_checkpoint": "beit-base-patch16-224-hasta-65-fold2/checkpoint-70", | |
| "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": 0.3055555555555556, | |
| "eval_loss": 1.3013406991958618, | |
| "eval_runtime": 0.5044, | |
| "eval_samples_per_second": 71.372, | |
| "eval_steps_per_second": 3.965, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "eval_accuracy": 0.2777777777777778, | |
| "eval_loss": 1.2798850536346436, | |
| "eval_runtime": 0.5041, | |
| "eval_samples_per_second": 71.421, | |
| "eval_steps_per_second": 3.968, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.3333333333333333, | |
| "eval_loss": 1.25881028175354, | |
| "eval_runtime": 0.5099, | |
| "eval_samples_per_second": 70.6, | |
| "eval_steps_per_second": 3.922, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.1295994520187378, | |
| "eval_runtime": 0.5017, | |
| "eval_samples_per_second": 71.749, | |
| "eval_steps_per_second": 3.986, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "eval_accuracy": 0.3611111111111111, | |
| "eval_loss": 1.1295137405395508, | |
| "eval_runtime": 0.5037, | |
| "eval_samples_per_second": 71.465, | |
| "eval_steps_per_second": 3.97, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 5.352811813354492, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1611, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "eval_accuracy": 0.25, | |
| "eval_loss": 1.2688605785369873, | |
| "eval_runtime": 0.4898, | |
| "eval_samples_per_second": 73.493, | |
| "eval_steps_per_second": 4.083, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.0894914865493774, | |
| "eval_runtime": 0.4962, | |
| "eval_samples_per_second": 72.556, | |
| "eval_steps_per_second": 4.031, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0977829694747925, | |
| "eval_runtime": 0.4938, | |
| "eval_samples_per_second": 72.909, | |
| "eval_steps_per_second": 4.05, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.1167902946472168, | |
| "eval_runtime": 0.4961, | |
| "eval_samples_per_second": 72.565, | |
| "eval_steps_per_second": 4.031, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.0844398736953735, | |
| "eval_runtime": 0.5056, | |
| "eval_samples_per_second": 71.205, | |
| "eval_steps_per_second": 3.956, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0475748777389526, | |
| "eval_runtime": 0.522, | |
| "eval_samples_per_second": 68.966, | |
| "eval_steps_per_second": 3.831, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 7.376298904418945, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.9913, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.2314592599868774, | |
| "eval_runtime": 0.4964, | |
| "eval_samples_per_second": 72.521, | |
| "eval_steps_per_second": 4.029, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.1443606615066528, | |
| "eval_runtime": 0.4944, | |
| "eval_samples_per_second": 72.822, | |
| "eval_steps_per_second": 4.046, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0241702795028687, | |
| "eval_runtime": 0.4918, | |
| "eval_samples_per_second": 73.196, | |
| "eval_steps_per_second": 4.066, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0495141744613647, | |
| "eval_runtime": 0.4977, | |
| "eval_samples_per_second": 72.33, | |
| "eval_steps_per_second": 4.018, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.1233758926391602, | |
| "eval_runtime": 0.4935, | |
| "eval_samples_per_second": 72.955, | |
| "eval_steps_per_second": 4.053, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 16.571428571428573, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.2331656217575073, | |
| "eval_runtime": 0.4988, | |
| "eval_samples_per_second": 72.17, | |
| "eval_steps_per_second": 4.009, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 8.922948837280273, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.9206, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 17.714285714285715, | |
| "eval_accuracy": 0.3611111111111111, | |
| "eval_loss": 1.4389086961746216, | |
| "eval_runtime": 0.5072, | |
| "eval_samples_per_second": 70.985, | |
| "eval_steps_per_second": 3.944, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.029965877532959, | |
| "eval_runtime": 0.5007, | |
| "eval_samples_per_second": 71.895, | |
| "eval_steps_per_second": 3.994, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0027683973312378, | |
| "eval_runtime": 0.4936, | |
| "eval_samples_per_second": 72.932, | |
| "eval_steps_per_second": 4.052, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 20.571428571428573, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0321506261825562, | |
| "eval_runtime": 0.5024, | |
| "eval_samples_per_second": 71.656, | |
| "eval_steps_per_second": 3.981, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 21.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0870566368103027, | |
| "eval_runtime": 0.5054, | |
| "eval_samples_per_second": 71.224, | |
| "eval_steps_per_second": 3.957, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 4.528126239776611, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.7309, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 0.9615763425827026, | |
| "eval_runtime": 0.5112, | |
| "eval_samples_per_second": 70.426, | |
| "eval_steps_per_second": 3.913, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9571282267570496, | |
| "eval_runtime": 0.4991, | |
| "eval_samples_per_second": 72.137, | |
| "eval_steps_per_second": 4.008, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 24.571428571428573, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9854538440704346, | |
| "eval_runtime": 0.5234, | |
| "eval_samples_per_second": 68.775, | |
| "eval_steps_per_second": 3.821, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9597752094268799, | |
| "eval_runtime": 0.4935, | |
| "eval_samples_per_second": 72.946, | |
| "eval_steps_per_second": 4.053, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.977421224117279, | |
| "eval_runtime": 0.4919, | |
| "eval_samples_per_second": 73.189, | |
| "eval_steps_per_second": 4.066, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9204927086830139, | |
| "eval_runtime": 0.4938, | |
| "eval_samples_per_second": 72.906, | |
| "eval_steps_per_second": 4.05, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 6.5547027587890625, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.6039, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.907256007194519, | |
| "eval_runtime": 0.4929, | |
| "eval_samples_per_second": 73.032, | |
| "eval_steps_per_second": 4.057, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 29.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8644082546234131, | |
| "eval_runtime": 0.5082, | |
| "eval_samples_per_second": 70.843, | |
| "eval_steps_per_second": 3.936, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.893060564994812, | |
| "eval_runtime": 0.4969, | |
| "eval_samples_per_second": 72.456, | |
| "eval_steps_per_second": 4.025, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.8686391115188599, | |
| "eval_runtime": 0.4943, | |
| "eval_samples_per_second": 72.823, | |
| "eval_steps_per_second": 4.046, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 32.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8380641937255859, | |
| "eval_runtime": 0.4948, | |
| "eval_samples_per_second": 72.763, | |
| "eval_steps_per_second": 4.042, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 33.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.8658040165901184, | |
| "eval_runtime": 0.5037, | |
| "eval_samples_per_second": 71.471, | |
| "eval_steps_per_second": 3.971, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 8.450542449951172, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.4784, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.991539716720581, | |
| "eval_runtime": 0.5076, | |
| "eval_samples_per_second": 70.918, | |
| "eval_steps_per_second": 3.94, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.7971311807632446, | |
| "eval_runtime": 0.5042, | |
| "eval_samples_per_second": 71.398, | |
| "eval_steps_per_second": 3.967, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 36.57142857142857, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.7682293653488159, | |
| "eval_runtime": 0.5041, | |
| "eval_samples_per_second": 71.418, | |
| "eval_steps_per_second": 3.968, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 37.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9360617399215698, | |
| "eval_runtime": 0.4962, | |
| "eval_samples_per_second": 72.553, | |
| "eval_steps_per_second": 4.031, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9093003869056702, | |
| "eval_runtime": 0.5003, | |
| "eval_samples_per_second": 71.951, | |
| "eval_steps_per_second": 3.997, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 7.291428089141846, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.4469, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.6727918982505798, | |
| "eval_runtime": 0.4995, | |
| "eval_samples_per_second": 72.069, | |
| "eval_steps_per_second": 4.004, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.57142857142857, | |
| "eval_accuracy": 0.7222222222222222, | |
| "eval_loss": 0.6414628028869629, | |
| "eval_runtime": 0.5093, | |
| "eval_samples_per_second": 70.688, | |
| "eval_steps_per_second": 3.927, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 41.714285714285715, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 0.7045332193374634, | |
| "eval_runtime": 0.4971, | |
| "eval_samples_per_second": 72.414, | |
| "eval_steps_per_second": 4.023, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.8973742723464966, | |
| "eval_runtime": 0.5013, | |
| "eval_samples_per_second": 71.811, | |
| "eval_steps_per_second": 3.99, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.8031913042068481, | |
| "eval_runtime": 0.5121, | |
| "eval_samples_per_second": 70.3, | |
| "eval_steps_per_second": 3.906, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 44.57142857142857, | |
| "eval_accuracy": 0.6944444444444444, | |
| "eval_loss": 0.7133803963661194, | |
| "eval_runtime": 0.5003, | |
| "eval_samples_per_second": 71.959, | |
| "eval_steps_per_second": 3.998, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 10.450151443481445, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.4329, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "eval_accuracy": 0.7222222222222222, | |
| "eval_loss": 0.6974515318870544, | |
| "eval_runtime": 0.519, | |
| "eval_samples_per_second": 69.368, | |
| "eval_steps_per_second": 3.854, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.7222222222222222, | |
| "eval_loss": 0.6758300065994263, | |
| "eval_runtime": 0.5051, | |
| "eval_samples_per_second": 71.277, | |
| "eval_steps_per_second": 3.96, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.8327140808105469, | |
| "eval_runtime": 0.4989, | |
| "eval_samples_per_second": 72.159, | |
| "eval_steps_per_second": 4.009, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9089304804801941, | |
| "eval_runtime": 0.49, | |
| "eval_samples_per_second": 73.464, | |
| "eval_steps_per_second": 4.081, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 49.714285714285715, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9157822132110596, | |
| "eval_runtime": 0.4989, | |
| "eval_samples_per_second": 72.158, | |
| "eval_steps_per_second": 4.009, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.8007399439811707, | |
| "eval_runtime": 0.502, | |
| "eval_samples_per_second": 71.715, | |
| "eval_steps_per_second": 3.984, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 5.677727222442627, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.4282, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.7362669110298157, | |
| "eval_runtime": 0.5031, | |
| "eval_samples_per_second": 71.553, | |
| "eval_steps_per_second": 3.975, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 52.57142857142857, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.737794816493988, | |
| "eval_runtime": 0.4965, | |
| "eval_samples_per_second": 72.504, | |
| "eval_steps_per_second": 4.028, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 53.714285714285715, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.7448822855949402, | |
| "eval_runtime": 0.494, | |
| "eval_samples_per_second": 72.869, | |
| "eval_steps_per_second": 4.048, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.7604518532752991, | |
| "eval_runtime": 0.5097, | |
| "eval_samples_per_second": 70.635, | |
| "eval_steps_per_second": 3.924, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.7852640151977539, | |
| "eval_runtime": 0.4974, | |
| "eval_samples_per_second": 72.373, | |
| "eval_steps_per_second": 4.021, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 56.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.7903004288673401, | |
| "eval_runtime": 0.4943, | |
| "eval_samples_per_second": 72.829, | |
| "eval_steps_per_second": 4.046, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 5.579706192016602, | |
| "learning_rate": 0.0, | |
| "loss": 0.3188, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.7925503849983215, | |
| "eval_runtime": 0.4996, | |
| "eval_samples_per_second": 72.055, | |
| "eval_steps_per_second": 4.003, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "step": 100, | |
| "total_flos": 8.899329213047992e+17, | |
| "train_loss": 0.6512869620323181, | |
| "train_runtime": 554.4555, | |
| "train_samples_per_second": 36.252, | |
| "train_steps_per_second": 0.18 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.6727918982505798, | |
| "eval_runtime": 0.5466, | |
| "eval_samples_per_second": 65.861, | |
| "eval_steps_per_second": 3.659, | |
| "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": 8.899329213047992e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |