| { | |
| "best_metric": 0.6944444444444444, | |
| "best_model_checkpoint": "beit-base-patch16-224-hasta-65-fold1/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": 0.3333333333333333, | |
| "eval_loss": 1.1951342821121216, | |
| "eval_runtime": 0.5089, | |
| "eval_samples_per_second": 70.74, | |
| "eval_steps_per_second": 3.93, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "eval_accuracy": 0.3055555555555556, | |
| "eval_loss": 1.2409210205078125, | |
| "eval_runtime": 0.5155, | |
| "eval_samples_per_second": 69.841, | |
| "eval_steps_per_second": 3.88, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.3055555555555556, | |
| "eval_loss": 1.218138337135315, | |
| "eval_runtime": 0.5198, | |
| "eval_samples_per_second": 69.261, | |
| "eval_steps_per_second": 3.848, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.1308220624923706, | |
| "eval_runtime": 0.5152, | |
| "eval_samples_per_second": 69.87, | |
| "eval_steps_per_second": 3.882, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.1857770681381226, | |
| "eval_runtime": 0.5007, | |
| "eval_samples_per_second": 71.904, | |
| "eval_steps_per_second": 3.995, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 4.952281475067139, | |
| "learning_rate": 5e-05, | |
| "loss": 1.092, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "eval_accuracy": 0.3055555555555556, | |
| "eval_loss": 1.2797729969024658, | |
| "eval_runtime": 0.5121, | |
| "eval_samples_per_second": 70.303, | |
| "eval_steps_per_second": 3.906, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.3611111111111111, | |
| "eval_loss": 1.0892032384872437, | |
| "eval_runtime": 0.5025, | |
| "eval_samples_per_second": 71.636, | |
| "eval_steps_per_second": 3.98, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.2658005952835083, | |
| "eval_runtime": 0.5197, | |
| "eval_samples_per_second": 69.272, | |
| "eval_steps_per_second": 3.848, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.1603169441223145, | |
| "eval_runtime": 0.5136, | |
| "eval_samples_per_second": 70.097, | |
| "eval_steps_per_second": 3.894, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.0786064863204956, | |
| "eval_runtime": 0.5166, | |
| "eval_samples_per_second": 69.688, | |
| "eval_steps_per_second": 3.872, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.2399791479110718, | |
| "eval_runtime": 0.5016, | |
| "eval_samples_per_second": 71.767, | |
| "eval_steps_per_second": 3.987, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 11.269621849060059, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.9727, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.292936086654663, | |
| "eval_runtime": 0.4771, | |
| "eval_samples_per_second": 75.46, | |
| "eval_steps_per_second": 4.192, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.161321997642517, | |
| "eval_runtime": 0.497, | |
| "eval_samples_per_second": 72.439, | |
| "eval_steps_per_second": 4.024, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9838488101959229, | |
| "eval_runtime": 0.4914, | |
| "eval_samples_per_second": 73.257, | |
| "eval_steps_per_second": 4.07, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.2316433191299438, | |
| "eval_runtime": 0.4927, | |
| "eval_samples_per_second": 73.074, | |
| "eval_steps_per_second": 4.06, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.151979923248291, | |
| "eval_runtime": 0.5114, | |
| "eval_samples_per_second": 70.402, | |
| "eval_steps_per_second": 3.911, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 16.571428571428573, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.034508466720581, | |
| "eval_runtime": 0.4983, | |
| "eval_samples_per_second": 72.244, | |
| "eval_steps_per_second": 4.014, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 8.270461082458496, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.8844, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 17.714285714285715, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.999970555305481, | |
| "eval_runtime": 0.5004, | |
| "eval_samples_per_second": 71.949, | |
| "eval_steps_per_second": 3.997, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9932595491409302, | |
| "eval_runtime": 0.506, | |
| "eval_samples_per_second": 71.149, | |
| "eval_steps_per_second": 3.953, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0511807203292847, | |
| "eval_runtime": 0.5024, | |
| "eval_samples_per_second": 71.658, | |
| "eval_steps_per_second": 3.981, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 20.571428571428573, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9949522018432617, | |
| "eval_runtime": 0.498, | |
| "eval_samples_per_second": 72.292, | |
| "eval_steps_per_second": 4.016, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 21.714285714285715, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 0.9621425271034241, | |
| "eval_runtime": 0.5011, | |
| "eval_samples_per_second": 71.845, | |
| "eval_steps_per_second": 3.991, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 9.892123222351074, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.7447, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.8811734318733215, | |
| "eval_runtime": 0.5013, | |
| "eval_samples_per_second": 71.816, | |
| "eval_steps_per_second": 3.99, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.024406909942627, | |
| "eval_runtime": 0.4975, | |
| "eval_samples_per_second": 72.357, | |
| "eval_steps_per_second": 4.02, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 24.571428571428573, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0124177932739258, | |
| "eval_runtime": 0.5032, | |
| "eval_samples_per_second": 71.543, | |
| "eval_steps_per_second": 3.975, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.8907694220542908, | |
| "eval_runtime": 0.501, | |
| "eval_samples_per_second": 71.85, | |
| "eval_steps_per_second": 3.992, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8184720873832703, | |
| "eval_runtime": 0.5096, | |
| "eval_samples_per_second": 70.643, | |
| "eval_steps_per_second": 3.925, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9408966898918152, | |
| "eval_runtime": 0.499, | |
| "eval_samples_per_second": 72.139, | |
| "eval_steps_per_second": 4.008, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 6.166210174560547, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.6176, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0401257276535034, | |
| "eval_runtime": 0.5125, | |
| "eval_samples_per_second": 70.249, | |
| "eval_steps_per_second": 3.903, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 29.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0988928079605103, | |
| "eval_runtime": 0.4975, | |
| "eval_samples_per_second": 72.356, | |
| "eval_steps_per_second": 4.02, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9102041721343994, | |
| "eval_runtime": 0.4988, | |
| "eval_samples_per_second": 72.176, | |
| "eval_steps_per_second": 4.01, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8855189681053162, | |
| "eval_runtime": 0.5026, | |
| "eval_samples_per_second": 71.634, | |
| "eval_steps_per_second": 3.98, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 32.57142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.8973506093025208, | |
| "eval_runtime": 0.4968, | |
| "eval_samples_per_second": 72.468, | |
| "eval_steps_per_second": 4.026, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 33.714285714285715, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9418532848358154, | |
| "eval_runtime": 0.4995, | |
| "eval_samples_per_second": 72.073, | |
| "eval_steps_per_second": 4.004, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 10.520373344421387, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.4929, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9471322298049927, | |
| "eval_runtime": 0.5068, | |
| "eval_samples_per_second": 71.029, | |
| "eval_steps_per_second": 3.946, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8608690500259399, | |
| "eval_runtime": 0.4977, | |
| "eval_samples_per_second": 72.332, | |
| "eval_steps_per_second": 4.018, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 36.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8558059334754944, | |
| "eval_runtime": 0.4937, | |
| "eval_samples_per_second": 72.92, | |
| "eval_steps_per_second": 4.051, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 37.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.8449121117591858, | |
| "eval_runtime": 0.5013, | |
| "eval_samples_per_second": 71.82, | |
| "eval_steps_per_second": 3.99, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 0.8136334419250488, | |
| "eval_runtime": 0.4944, | |
| "eval_samples_per_second": 72.811, | |
| "eval_steps_per_second": 4.045, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 11.665092468261719, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.463, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.6944444444444444, | |
| "eval_loss": 0.8281329870223999, | |
| "eval_runtime": 0.5025, | |
| "eval_samples_per_second": 71.639, | |
| "eval_steps_per_second": 3.98, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.57142857142857, | |
| "eval_accuracy": 0.6944444444444444, | |
| "eval_loss": 0.8226919770240784, | |
| "eval_runtime": 0.5104, | |
| "eval_samples_per_second": 70.529, | |
| "eval_steps_per_second": 3.918, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 41.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8322837352752686, | |
| "eval_runtime": 0.4973, | |
| "eval_samples_per_second": 72.392, | |
| "eval_steps_per_second": 4.022, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8435983657836914, | |
| "eval_runtime": 0.4888, | |
| "eval_samples_per_second": 73.644, | |
| "eval_steps_per_second": 4.091, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.839017927646637, | |
| "eval_runtime": 0.4935, | |
| "eval_samples_per_second": 72.953, | |
| "eval_steps_per_second": 4.053, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 44.57142857142857, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.8580043911933899, | |
| "eval_runtime": 0.4996, | |
| "eval_samples_per_second": 72.055, | |
| "eval_steps_per_second": 4.003, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 5.9656195640563965, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.3995, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9374995827674866, | |
| "eval_runtime": 0.5119, | |
| "eval_samples_per_second": 70.321, | |
| "eval_steps_per_second": 3.907, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9897011518478394, | |
| "eval_runtime": 0.5037, | |
| "eval_samples_per_second": 71.47, | |
| "eval_steps_per_second": 3.971, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9784960150718689, | |
| "eval_runtime": 0.4971, | |
| "eval_samples_per_second": 72.425, | |
| "eval_steps_per_second": 4.024, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9335712790489197, | |
| "eval_runtime": 0.5144, | |
| "eval_samples_per_second": 69.989, | |
| "eval_steps_per_second": 3.888, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 49.714285714285715, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.8504276275634766, | |
| "eval_runtime": 0.5039, | |
| "eval_samples_per_second": 71.439, | |
| "eval_steps_per_second": 3.969, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 0.8450478911399841, | |
| "eval_runtime": 0.5024, | |
| "eval_samples_per_second": 71.65, | |
| "eval_steps_per_second": 3.981, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 10.226240158081055, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.3697, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.8530892133712769, | |
| "eval_runtime": 0.5051, | |
| "eval_samples_per_second": 71.274, | |
| "eval_steps_per_second": 3.96, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 52.57142857142857, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.8728136420249939, | |
| "eval_runtime": 0.5023, | |
| "eval_samples_per_second": 71.671, | |
| "eval_steps_per_second": 3.982, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 53.714285714285715, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 0.9075747728347778, | |
| "eval_runtime": 0.4997, | |
| "eval_samples_per_second": 72.043, | |
| "eval_steps_per_second": 4.002, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9174892902374268, | |
| "eval_runtime": 0.5088, | |
| "eval_samples_per_second": 70.753, | |
| "eval_steps_per_second": 3.931, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9145221710205078, | |
| "eval_runtime": 0.5129, | |
| "eval_samples_per_second": 70.19, | |
| "eval_steps_per_second": 3.899, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 56.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9118825197219849, | |
| "eval_runtime": 0.4996, | |
| "eval_samples_per_second": 72.056, | |
| "eval_steps_per_second": 4.003, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 6.797101020812988, | |
| "learning_rate": 0.0, | |
| "loss": 0.3259, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9102092385292053, | |
| "eval_runtime": 0.5165, | |
| "eval_samples_per_second": 69.7, | |
| "eval_steps_per_second": 3.872, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "step": 100, | |
| "total_flos": 8.899329213047992e+17, | |
| "train_loss": 0.6362448835372925, | |
| "train_runtime": 561.1572, | |
| "train_samples_per_second": 35.819, | |
| "train_steps_per_second": 0.178 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.6944444444444444, | |
| "eval_loss": 0.8281329870223999, | |
| "eval_runtime": 0.5522, | |
| "eval_samples_per_second": 65.19, | |
| "eval_steps_per_second": 3.622, | |
| "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 | |
| } | |