| { | |
| "best_metric": 0.6388888888888888, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-hasta-65-fold1/checkpoint-42", | |
| "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": 0.2222222222222222, | |
| "eval_loss": 1.3178210258483887, | |
| "eval_runtime": 0.4939, | |
| "eval_samples_per_second": 72.895, | |
| "eval_steps_per_second": 4.05, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "eval_accuracy": 0.2777777777777778, | |
| "eval_loss": 1.4013595581054688, | |
| "eval_runtime": 0.4989, | |
| "eval_samples_per_second": 72.154, | |
| "eval_steps_per_second": 4.009, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.2777777777777778, | |
| "eval_loss": 1.353532075881958, | |
| "eval_runtime": 0.5077, | |
| "eval_samples_per_second": 70.908, | |
| "eval_steps_per_second": 3.939, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3055555555555556, | |
| "eval_loss": 1.1299159526824951, | |
| "eval_runtime": 0.529, | |
| "eval_samples_per_second": 68.047, | |
| "eval_steps_per_second": 3.78, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.0860486030578613, | |
| "eval_runtime": 0.5016, | |
| "eval_samples_per_second": 71.764, | |
| "eval_steps_per_second": 3.987, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 3.6564159393310547, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0868, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "eval_accuracy": 0.3333333333333333, | |
| "eval_loss": 1.1120641231536865, | |
| "eval_runtime": 0.4949, | |
| "eval_samples_per_second": 72.737, | |
| "eval_steps_per_second": 4.041, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.3611111111111111, | |
| "eval_loss": 1.0690841674804688, | |
| "eval_runtime": 0.4834, | |
| "eval_samples_per_second": 74.47, | |
| "eval_steps_per_second": 4.137, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0270113945007324, | |
| "eval_runtime": 0.4858, | |
| "eval_samples_per_second": 74.1, | |
| "eval_steps_per_second": 4.117, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0360279083251953, | |
| "eval_runtime": 0.4936, | |
| "eval_samples_per_second": 72.928, | |
| "eval_steps_per_second": 4.052, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.0385184288024902, | |
| "eval_runtime": 0.4841, | |
| "eval_samples_per_second": 74.37, | |
| "eval_steps_per_second": 4.132, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 0.9951233267784119, | |
| "eval_runtime": 0.4898, | |
| "eval_samples_per_second": 73.501, | |
| "eval_steps_per_second": 4.083, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 4.391091823577881, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.9487, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.0028576850891113, | |
| "eval_runtime": 0.4991, | |
| "eval_samples_per_second": 72.133, | |
| "eval_steps_per_second": 4.007, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.0133533477783203, | |
| "eval_runtime": 0.5002, | |
| "eval_samples_per_second": 71.971, | |
| "eval_steps_per_second": 3.998, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 0.9598518013954163, | |
| "eval_runtime": 0.4936, | |
| "eval_samples_per_second": 72.926, | |
| "eval_steps_per_second": 4.051, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9116785526275635, | |
| "eval_runtime": 0.489, | |
| "eval_samples_per_second": 73.626, | |
| "eval_steps_per_second": 4.09, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.8856263160705566, | |
| "eval_runtime": 0.4884, | |
| "eval_samples_per_second": 73.703, | |
| "eval_steps_per_second": 4.095, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 16.571428571428573, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 0.9275122880935669, | |
| "eval_runtime": 0.4919, | |
| "eval_samples_per_second": 73.18, | |
| "eval_steps_per_second": 4.066, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 6.513401031494141, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.7942, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 17.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.904094934463501, | |
| "eval_runtime": 0.4892, | |
| "eval_samples_per_second": 73.594, | |
| "eval_steps_per_second": 4.089, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 0.8999477028846741, | |
| "eval_runtime": 0.5067, | |
| "eval_samples_per_second": 71.051, | |
| "eval_steps_per_second": 3.947, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8831884264945984, | |
| "eval_runtime": 0.4943, | |
| "eval_samples_per_second": 72.834, | |
| "eval_steps_per_second": 4.046, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 20.571428571428573, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.886363685131073, | |
| "eval_runtime": 0.4873, | |
| "eval_samples_per_second": 73.873, | |
| "eval_steps_per_second": 4.104, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 21.714285714285715, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.8550977110862732, | |
| "eval_runtime": 0.5223, | |
| "eval_samples_per_second": 68.925, | |
| "eval_steps_per_second": 3.829, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 8.051746368408203, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.5911, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.8241526484489441, | |
| "eval_runtime": 0.4922, | |
| "eval_samples_per_second": 73.144, | |
| "eval_steps_per_second": 4.064, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9265490770339966, | |
| "eval_runtime": 0.4936, | |
| "eval_samples_per_second": 72.931, | |
| "eval_steps_per_second": 4.052, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 24.571428571428573, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.867362916469574, | |
| "eval_runtime": 0.494, | |
| "eval_samples_per_second": 72.868, | |
| "eval_steps_per_second": 4.048, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.7891539335250854, | |
| "eval_runtime": 0.4871, | |
| "eval_samples_per_second": 73.904, | |
| "eval_steps_per_second": 4.106, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.800545334815979, | |
| "eval_runtime": 0.5014, | |
| "eval_samples_per_second": 71.801, | |
| "eval_steps_per_second": 3.989, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.830207884311676, | |
| "eval_runtime": 0.4985, | |
| "eval_samples_per_second": 72.214, | |
| "eval_steps_per_second": 4.012, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 3.0137877464294434, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.4865, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.8892796039581299, | |
| "eval_runtime": 0.4944, | |
| "eval_samples_per_second": 72.808, | |
| "eval_steps_per_second": 4.045, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 29.714285714285715, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9043471813201904, | |
| "eval_runtime": 0.5016, | |
| "eval_samples_per_second": 71.776, | |
| "eval_steps_per_second": 3.988, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8433141112327576, | |
| "eval_runtime": 0.4867, | |
| "eval_samples_per_second": 73.965, | |
| "eval_steps_per_second": 4.109, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8676599860191345, | |
| "eval_runtime": 0.4892, | |
| "eval_samples_per_second": 73.589, | |
| "eval_steps_per_second": 4.088, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 32.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9008312225341797, | |
| "eval_runtime": 0.496, | |
| "eval_samples_per_second": 72.575, | |
| "eval_steps_per_second": 4.032, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 33.714285714285715, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.953251302242279, | |
| "eval_runtime": 0.4918, | |
| "eval_samples_per_second": 73.203, | |
| "eval_steps_per_second": 4.067, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 5.712308883666992, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.4007, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9174816608428955, | |
| "eval_runtime": 0.4894, | |
| "eval_samples_per_second": 73.553, | |
| "eval_steps_per_second": 4.086, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9090279340744019, | |
| "eval_runtime": 0.507, | |
| "eval_samples_per_second": 71.007, | |
| "eval_steps_per_second": 3.945, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 36.57142857142857, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0004390478134155, | |
| "eval_runtime": 0.4948, | |
| "eval_samples_per_second": 72.751, | |
| "eval_steps_per_second": 4.042, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 37.714285714285715, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.039277195930481, | |
| "eval_runtime": 0.492, | |
| "eval_samples_per_second": 73.172, | |
| "eval_steps_per_second": 4.065, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9195742011070251, | |
| "eval_runtime": 0.4866, | |
| "eval_samples_per_second": 73.989, | |
| "eval_steps_per_second": 4.111, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 5.453234672546387, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.3691, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9504802227020264, | |
| "eval_runtime": 0.4966, | |
| "eval_samples_per_second": 72.495, | |
| "eval_steps_per_second": 4.027, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.57142857142857, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9633702039718628, | |
| "eval_runtime": 0.4918, | |
| "eval_samples_per_second": 73.206, | |
| "eval_steps_per_second": 4.067, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 41.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9718315601348877, | |
| "eval_runtime": 0.4916, | |
| "eval_samples_per_second": 73.233, | |
| "eval_steps_per_second": 4.069, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9256722331047058, | |
| "eval_runtime": 0.5113, | |
| "eval_samples_per_second": 70.413, | |
| "eval_steps_per_second": 3.912, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.901976466178894, | |
| "eval_runtime": 0.4898, | |
| "eval_samples_per_second": 73.496, | |
| "eval_steps_per_second": 4.083, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 44.57142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9132314324378967, | |
| "eval_runtime": 0.5127, | |
| "eval_samples_per_second": 70.223, | |
| "eval_steps_per_second": 3.901, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 2.8120012283325195, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.3278, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0339668989181519, | |
| "eval_runtime": 0.4996, | |
| "eval_samples_per_second": 72.051, | |
| "eval_steps_per_second": 4.003, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0932819843292236, | |
| "eval_runtime": 0.4949, | |
| "eval_samples_per_second": 72.738, | |
| "eval_steps_per_second": 4.041, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.023079514503479, | |
| "eval_runtime": 0.4983, | |
| "eval_samples_per_second": 72.253, | |
| "eval_steps_per_second": 4.014, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9826453328132629, | |
| "eval_runtime": 0.5029, | |
| "eval_samples_per_second": 71.591, | |
| "eval_steps_per_second": 3.977, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 49.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9328654408454895, | |
| "eval_runtime": 0.485, | |
| "eval_samples_per_second": 74.233, | |
| "eval_steps_per_second": 4.124, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9280245304107666, | |
| "eval_runtime": 0.483, | |
| "eval_samples_per_second": 74.54, | |
| "eval_steps_per_second": 4.141, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 3.2972934246063232, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.2909, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9311569929122925, | |
| "eval_runtime": 0.4834, | |
| "eval_samples_per_second": 74.47, | |
| "eval_steps_per_second": 4.137, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 52.57142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9359264373779297, | |
| "eval_runtime": 0.4905, | |
| "eval_samples_per_second": 73.399, | |
| "eval_steps_per_second": 4.078, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 53.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.949504017829895, | |
| "eval_runtime": 0.4958, | |
| "eval_samples_per_second": 72.616, | |
| "eval_steps_per_second": 4.034, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9607170820236206, | |
| "eval_runtime": 0.5024, | |
| "eval_samples_per_second": 71.657, | |
| "eval_steps_per_second": 3.981, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9684595465660095, | |
| "eval_runtime": 0.4974, | |
| "eval_samples_per_second": 72.381, | |
| "eval_steps_per_second": 4.021, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 56.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.970331072807312, | |
| "eval_runtime": 0.4957, | |
| "eval_samples_per_second": 72.62, | |
| "eval_steps_per_second": 4.034, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 4.024988174438477, | |
| "learning_rate": 0.0, | |
| "loss": 0.2697, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9713256359100342, | |
| "eval_runtime": 0.4843, | |
| "eval_samples_per_second": 74.333, | |
| "eval_steps_per_second": 4.13, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "step": 100, | |
| "total_flos": 8.903293865079828e+17, | |
| "train_loss": 0.5565648937225341, | |
| "train_runtime": 558.9878, | |
| "train_samples_per_second": 35.958, | |
| "train_steps_per_second": 0.179 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9265490770339966, | |
| "eval_runtime": 0.5552, | |
| "eval_samples_per_second": 64.836, | |
| "eval_steps_per_second": 3.602, | |
| "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": 8.903293865079828e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |