| { | |
| "best_metric": 0.6666666666666666, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-hasta-55-fold2/checkpoint-89", | |
| "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": 0.3611111111111111, | |
| "eval_loss": 1.3240900039672852, | |
| "eval_runtime": 0.5104, | |
| "eval_samples_per_second": 70.53, | |
| "eval_steps_per_second": 3.918, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "eval_accuracy": 0.3055555555555556, | |
| "eval_loss": 1.1597557067871094, | |
| "eval_runtime": 0.5022, | |
| "eval_samples_per_second": 71.682, | |
| "eval_steps_per_second": 3.982, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.192773699760437, | |
| "eval_runtime": 0.5084, | |
| "eval_samples_per_second": 70.809, | |
| "eval_steps_per_second": 3.934, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3333333333333333, | |
| "eval_loss": 1.1229112148284912, | |
| "eval_runtime": 0.5131, | |
| "eval_samples_per_second": 70.162, | |
| "eval_steps_per_second": 3.898, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "eval_accuracy": 0.3333333333333333, | |
| "eval_loss": 1.111149787902832, | |
| "eval_runtime": 0.5069, | |
| "eval_samples_per_second": 71.02, | |
| "eval_steps_per_second": 3.946, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 2.360438108444214, | |
| "learning_rate": 5e-05, | |
| "loss": 1.13, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "eval_accuracy": 0.3611111111111111, | |
| "eval_loss": 1.1062475442886353, | |
| "eval_runtime": 0.493, | |
| "eval_samples_per_second": 73.017, | |
| "eval_steps_per_second": 4.056, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.0655484199523926, | |
| "eval_runtime": 0.4943, | |
| "eval_samples_per_second": 72.825, | |
| "eval_steps_per_second": 4.046, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.0526686906814575, | |
| "eval_runtime": 0.5085, | |
| "eval_samples_per_second": 70.799, | |
| "eval_steps_per_second": 3.933, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.0448135137557983, | |
| "eval_runtime": 0.4842, | |
| "eval_samples_per_second": 74.357, | |
| "eval_steps_per_second": 4.131, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9955484867095947, | |
| "eval_runtime": 0.4943, | |
| "eval_samples_per_second": 72.838, | |
| "eval_steps_per_second": 4.047, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9818601608276367, | |
| "eval_runtime": 0.4991, | |
| "eval_samples_per_second": 72.131, | |
| "eval_steps_per_second": 4.007, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 4.555427074432373, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.9629, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.9722151160240173, | |
| "eval_runtime": 0.515, | |
| "eval_samples_per_second": 69.903, | |
| "eval_steps_per_second": 3.884, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 0.9777337312698364, | |
| "eval_runtime": 0.5112, | |
| "eval_samples_per_second": 70.422, | |
| "eval_steps_per_second": 3.912, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9125936031341553, | |
| "eval_runtime": 0.5049, | |
| "eval_samples_per_second": 71.306, | |
| "eval_steps_per_second": 3.961, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9931738972663879, | |
| "eval_runtime": 0.4932, | |
| "eval_samples_per_second": 72.999, | |
| "eval_steps_per_second": 4.055, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9945557713508606, | |
| "eval_runtime": 0.4975, | |
| "eval_samples_per_second": 72.361, | |
| "eval_steps_per_second": 4.02, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 16.571428571428573, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9689527153968811, | |
| "eval_runtime": 0.4915, | |
| "eval_samples_per_second": 73.247, | |
| "eval_steps_per_second": 4.069, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 5.037437915802002, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.822, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 17.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9163329005241394, | |
| "eval_runtime": 0.5011, | |
| "eval_samples_per_second": 71.847, | |
| "eval_steps_per_second": 3.992, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9798648357391357, | |
| "eval_runtime": 0.5146, | |
| "eval_samples_per_second": 69.964, | |
| "eval_steps_per_second": 3.887, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.927280843257904, | |
| "eval_runtime": 0.5069, | |
| "eval_samples_per_second": 71.026, | |
| "eval_steps_per_second": 3.946, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 20.571428571428573, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9007765650749207, | |
| "eval_runtime": 0.5031, | |
| "eval_samples_per_second": 71.557, | |
| "eval_steps_per_second": 3.975, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 21.714285714285715, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0304814577102661, | |
| "eval_runtime": 0.5055, | |
| "eval_samples_per_second": 71.219, | |
| "eval_steps_per_second": 3.957, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 4.266380786895752, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.7066, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9330142140388489, | |
| "eval_runtime": 0.4996, | |
| "eval_samples_per_second": 72.058, | |
| "eval_steps_per_second": 4.003, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9151402711868286, | |
| "eval_runtime": 0.496, | |
| "eval_samples_per_second": 72.587, | |
| "eval_steps_per_second": 4.033, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 24.571428571428573, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9213579893112183, | |
| "eval_runtime": 0.5032, | |
| "eval_samples_per_second": 71.546, | |
| "eval_steps_per_second": 3.975, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9322340488433838, | |
| "eval_runtime": 0.5005, | |
| "eval_samples_per_second": 71.923, | |
| "eval_steps_per_second": 3.996, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0533533096313477, | |
| "eval_runtime": 0.5031, | |
| "eval_samples_per_second": 71.554, | |
| "eval_steps_per_second": 3.975, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.1222937107086182, | |
| "eval_runtime": 0.5022, | |
| "eval_samples_per_second": 71.683, | |
| "eval_steps_per_second": 3.982, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 4.958626747131348, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.5702, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0080817937850952, | |
| "eval_runtime": 0.5117, | |
| "eval_samples_per_second": 70.357, | |
| "eval_steps_per_second": 3.909, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 29.714285714285715, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.8680193424224854, | |
| "eval_runtime": 0.5019, | |
| "eval_samples_per_second": 71.721, | |
| "eval_steps_per_second": 3.984, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9258506298065186, | |
| "eval_runtime": 0.5016, | |
| "eval_samples_per_second": 71.769, | |
| "eval_steps_per_second": 3.987, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9936321973800659, | |
| "eval_runtime": 0.4931, | |
| "eval_samples_per_second": 73.009, | |
| "eval_steps_per_second": 4.056, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 32.57142857142857, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.97624272108078, | |
| "eval_runtime": 0.4976, | |
| "eval_samples_per_second": 72.352, | |
| "eval_steps_per_second": 4.02, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 33.714285714285715, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9298437833786011, | |
| "eval_runtime": 0.5064, | |
| "eval_samples_per_second": 71.096, | |
| "eval_steps_per_second": 3.95, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 3.622584342956543, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.4903, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9352247714996338, | |
| "eval_runtime": 0.5084, | |
| "eval_samples_per_second": 70.807, | |
| "eval_steps_per_second": 3.934, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9918670654296875, | |
| "eval_runtime": 0.4987, | |
| "eval_samples_per_second": 72.181, | |
| "eval_steps_per_second": 4.01, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 36.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9660681486129761, | |
| "eval_runtime": 0.4985, | |
| "eval_samples_per_second": 72.222, | |
| "eval_steps_per_second": 4.012, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 37.714285714285715, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9763928651809692, | |
| "eval_runtime": 0.4967, | |
| "eval_samples_per_second": 72.481, | |
| "eval_steps_per_second": 4.027, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9909236431121826, | |
| "eval_runtime": 0.5073, | |
| "eval_samples_per_second": 70.958, | |
| "eval_steps_per_second": 3.942, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 5.279661178588867, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.3959, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9905844926834106, | |
| "eval_runtime": 0.5026, | |
| "eval_samples_per_second": 71.632, | |
| "eval_steps_per_second": 3.98, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.57142857142857, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9788994193077087, | |
| "eval_runtime": 0.5022, | |
| "eval_samples_per_second": 71.686, | |
| "eval_steps_per_second": 3.983, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 41.714285714285715, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9528616666793823, | |
| "eval_runtime": 0.5003, | |
| "eval_samples_per_second": 71.955, | |
| "eval_steps_per_second": 3.998, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9729149341583252, | |
| "eval_runtime": 0.4998, | |
| "eval_samples_per_second": 72.031, | |
| "eval_steps_per_second": 4.002, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.0425068140029907, | |
| "eval_runtime": 0.5045, | |
| "eval_samples_per_second": 71.364, | |
| "eval_steps_per_second": 3.965, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 44.57142857142857, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.0531220436096191, | |
| "eval_runtime": 0.5005, | |
| "eval_samples_per_second": 71.935, | |
| "eval_steps_per_second": 3.996, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 4.175187110900879, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.3792, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9799526929855347, | |
| "eval_runtime": 0.5041, | |
| "eval_samples_per_second": 71.411, | |
| "eval_steps_per_second": 3.967, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9608871936798096, | |
| "eval_runtime": 0.4995, | |
| "eval_samples_per_second": 72.067, | |
| "eval_steps_per_second": 4.004, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9819880723953247, | |
| "eval_runtime": 0.5009, | |
| "eval_samples_per_second": 71.877, | |
| "eval_steps_per_second": 3.993, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 1.025152564048767, | |
| "eval_runtime": 0.5001, | |
| "eval_samples_per_second": 71.982, | |
| "eval_steps_per_second": 3.999, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 49.714285714285715, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.0591676235198975, | |
| "eval_runtime": 0.498, | |
| "eval_samples_per_second": 72.294, | |
| "eval_steps_per_second": 4.016, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 1.023190975189209, | |
| "eval_runtime": 0.5158, | |
| "eval_samples_per_second": 69.8, | |
| "eval_steps_per_second": 3.878, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 3.914527416229248, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.3267, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9972048997879028, | |
| "eval_runtime": 0.4946, | |
| "eval_samples_per_second": 72.792, | |
| "eval_steps_per_second": 4.044, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 52.57142857142857, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9925190806388855, | |
| "eval_runtime": 0.4949, | |
| "eval_samples_per_second": 72.735, | |
| "eval_steps_per_second": 4.041, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 53.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.000749945640564, | |
| "eval_runtime": 0.5035, | |
| "eval_samples_per_second": 71.494, | |
| "eval_steps_per_second": 3.972, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0168888568878174, | |
| "eval_runtime": 0.5013, | |
| "eval_samples_per_second": 71.811, | |
| "eval_steps_per_second": 3.989, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0303747653961182, | |
| "eval_runtime": 0.4993, | |
| "eval_samples_per_second": 72.106, | |
| "eval_steps_per_second": 4.006, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 56.57142857142857, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.035979151725769, | |
| "eval_runtime": 0.4966, | |
| "eval_samples_per_second": 72.486, | |
| "eval_steps_per_second": 4.027, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 5.318258762359619, | |
| "learning_rate": 0.0, | |
| "loss": 0.2959, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.0385332107543945, | |
| "eval_runtime": 0.5052, | |
| "eval_samples_per_second": 71.254, | |
| "eval_steps_per_second": 3.959, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "step": 100, | |
| "total_flos": 8.770779002957046e+17, | |
| "train_loss": 0.6079896140098572, | |
| "train_runtime": 548.0225, | |
| "train_samples_per_second": 36.13, | |
| "train_steps_per_second": 0.182 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 1.023190975189209, | |
| "eval_runtime": 0.5699, | |
| "eval_samples_per_second": 63.166, | |
| "eval_steps_per_second": 3.509, | |
| "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 | |
| } | |