| { | |
| "best_metric": 0.6944444444444444, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-hasta-55-fold1/checkpoint-26", | |
| "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": 0.4444444444444444, | |
| "eval_loss": 1.1037373542785645, | |
| "eval_runtime": 0.4817, | |
| "eval_samples_per_second": 74.733, | |
| "eval_steps_per_second": 4.152, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.0649282932281494, | |
| "eval_runtime": 0.4961, | |
| "eval_samples_per_second": 72.566, | |
| "eval_steps_per_second": 4.031, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.072746753692627, | |
| "eval_runtime": 0.4913, | |
| "eval_samples_per_second": 73.282, | |
| "eval_steps_per_second": 4.071, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0622544288635254, | |
| "eval_runtime": 0.5294, | |
| "eval_samples_per_second": 68.005, | |
| "eval_steps_per_second": 3.778, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.0530375242233276, | |
| "eval_runtime": 0.5115, | |
| "eval_samples_per_second": 70.381, | |
| "eval_steps_per_second": 3.91, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 2.886908769607544, | |
| "learning_rate": 5e-05, | |
| "loss": 1.065, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0509940385818481, | |
| "eval_runtime": 0.5018, | |
| "eval_samples_per_second": 71.743, | |
| "eval_steps_per_second": 3.986, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.0744811296463013, | |
| "eval_runtime": 0.4917, | |
| "eval_samples_per_second": 73.209, | |
| "eval_steps_per_second": 4.067, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.05109441280365, | |
| "eval_runtime": 0.508, | |
| "eval_samples_per_second": 70.869, | |
| "eval_steps_per_second": 3.937, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0157655477523804, | |
| "eval_runtime": 0.495, | |
| "eval_samples_per_second": 72.732, | |
| "eval_steps_per_second": 4.041, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9998018741607666, | |
| "eval_runtime": 0.4874, | |
| "eval_samples_per_second": 73.856, | |
| "eval_steps_per_second": 4.103, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0471947193145752, | |
| "eval_runtime": 0.4963, | |
| "eval_samples_per_second": 72.536, | |
| "eval_steps_per_second": 4.03, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 4.345711708068848, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.9239, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9674879908561707, | |
| "eval_runtime": 0.4968, | |
| "eval_samples_per_second": 72.459, | |
| "eval_steps_per_second": 4.025, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9731967449188232, | |
| "eval_runtime": 0.5023, | |
| "eval_samples_per_second": 71.675, | |
| "eval_steps_per_second": 3.982, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.9489168524742126, | |
| "eval_runtime": 0.4888, | |
| "eval_samples_per_second": 73.643, | |
| "eval_steps_per_second": 4.091, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.6944444444444444, | |
| "eval_loss": 0.9276576638221741, | |
| "eval_runtime": 0.4932, | |
| "eval_samples_per_second": 72.996, | |
| "eval_steps_per_second": 4.055, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9243949055671692, | |
| "eval_runtime": 0.4933, | |
| "eval_samples_per_second": 72.984, | |
| "eval_steps_per_second": 4.055, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 16.571428571428573, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.964252769947052, | |
| "eval_runtime": 0.4952, | |
| "eval_samples_per_second": 72.703, | |
| "eval_steps_per_second": 4.039, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 3.782306432723999, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.7838, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 17.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9721066951751709, | |
| "eval_runtime": 0.5046, | |
| "eval_samples_per_second": 71.35, | |
| "eval_steps_per_second": 3.964, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9432339072227478, | |
| "eval_runtime": 0.502, | |
| "eval_samples_per_second": 71.712, | |
| "eval_steps_per_second": 3.984, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9336562752723694, | |
| "eval_runtime": 0.4955, | |
| "eval_samples_per_second": 72.651, | |
| "eval_steps_per_second": 4.036, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 20.571428571428573, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9368643164634705, | |
| "eval_runtime": 0.4873, | |
| "eval_samples_per_second": 73.884, | |
| "eval_steps_per_second": 4.105, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 21.714285714285715, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9481333494186401, | |
| "eval_runtime": 0.4924, | |
| "eval_samples_per_second": 73.112, | |
| "eval_steps_per_second": 4.062, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 5.4750075340271, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6124, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9642544984817505, | |
| "eval_runtime": 0.4904, | |
| "eval_samples_per_second": 73.406, | |
| "eval_steps_per_second": 4.078, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9299950003623962, | |
| "eval_runtime": 0.5066, | |
| "eval_samples_per_second": 71.062, | |
| "eval_steps_per_second": 3.948, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 24.571428571428573, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 0.9628922343254089, | |
| "eval_runtime": 0.4946, | |
| "eval_samples_per_second": 72.788, | |
| "eval_steps_per_second": 4.044, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9200094938278198, | |
| "eval_runtime": 0.5146, | |
| "eval_samples_per_second": 69.962, | |
| "eval_steps_per_second": 3.887, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0088926553726196, | |
| "eval_runtime": 0.4921, | |
| "eval_samples_per_second": 73.154, | |
| "eval_steps_per_second": 4.064, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9979636073112488, | |
| "eval_runtime": 0.4918, | |
| "eval_samples_per_second": 73.206, | |
| "eval_steps_per_second": 4.067, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 5.6271796226501465, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.5068, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9875730872154236, | |
| "eval_runtime": 0.4927, | |
| "eval_samples_per_second": 73.068, | |
| "eval_steps_per_second": 4.059, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 29.714285714285715, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0463812351226807, | |
| "eval_runtime": 0.4966, | |
| "eval_samples_per_second": 72.495, | |
| "eval_steps_per_second": 4.027, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0311660766601562, | |
| "eval_runtime": 0.4913, | |
| "eval_samples_per_second": 73.277, | |
| "eval_steps_per_second": 4.071, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0946307182312012, | |
| "eval_runtime": 0.495, | |
| "eval_samples_per_second": 72.731, | |
| "eval_steps_per_second": 4.041, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 32.57142857142857, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.1676592826843262, | |
| "eval_runtime": 0.5005, | |
| "eval_samples_per_second": 71.929, | |
| "eval_steps_per_second": 3.996, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 33.714285714285715, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.1615511178970337, | |
| "eval_runtime": 0.4916, | |
| "eval_samples_per_second": 73.227, | |
| "eval_steps_per_second": 4.068, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 6.785258769989014, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.4375, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0658056735992432, | |
| "eval_runtime": 0.5148, | |
| "eval_samples_per_second": 69.929, | |
| "eval_steps_per_second": 3.885, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.092119574546814, | |
| "eval_runtime": 0.4917, | |
| "eval_samples_per_second": 73.216, | |
| "eval_steps_per_second": 4.068, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 36.57142857142857, | |
| "eval_accuracy": 0.6388888888888888, | |
| "eval_loss": 1.0800608396530151, | |
| "eval_runtime": 0.4952, | |
| "eval_samples_per_second": 72.699, | |
| "eval_steps_per_second": 4.039, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 37.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0586397647857666, | |
| "eval_runtime": 0.4991, | |
| "eval_samples_per_second": 72.134, | |
| "eval_steps_per_second": 4.007, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.215223789215088, | |
| "eval_runtime": 0.5061, | |
| "eval_samples_per_second": 71.136, | |
| "eval_steps_per_second": 3.952, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 5.2041168212890625, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.3932, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.154255747795105, | |
| "eval_runtime": 0.5115, | |
| "eval_samples_per_second": 70.382, | |
| "eval_steps_per_second": 3.91, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.57142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0654683113098145, | |
| "eval_runtime": 0.5011, | |
| "eval_samples_per_second": 71.848, | |
| "eval_steps_per_second": 3.992, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 41.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.99521803855896, | |
| "eval_runtime": 0.4942, | |
| "eval_samples_per_second": 72.839, | |
| "eval_steps_per_second": 4.047, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9986051917076111, | |
| "eval_runtime": 0.5007, | |
| "eval_samples_per_second": 71.905, | |
| "eval_steps_per_second": 3.995, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0175378322601318, | |
| "eval_runtime": 0.4946, | |
| "eval_samples_per_second": 72.79, | |
| "eval_steps_per_second": 4.044, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 44.57142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0233629941940308, | |
| "eval_runtime": 0.5225, | |
| "eval_samples_per_second": 68.901, | |
| "eval_steps_per_second": 3.828, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 3.7592148780822754, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.3539, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0385427474975586, | |
| "eval_runtime": 0.4956, | |
| "eval_samples_per_second": 72.634, | |
| "eval_steps_per_second": 4.035, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0191251039505005, | |
| "eval_runtime": 0.5016, | |
| "eval_samples_per_second": 71.772, | |
| "eval_steps_per_second": 3.987, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0150564908981323, | |
| "eval_runtime": 0.4942, | |
| "eval_samples_per_second": 72.845, | |
| "eval_steps_per_second": 4.047, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0203356742858887, | |
| "eval_runtime": 0.4974, | |
| "eval_samples_per_second": 72.378, | |
| "eval_steps_per_second": 4.021, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 49.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.034138560295105, | |
| "eval_runtime": 0.5058, | |
| "eval_samples_per_second": 71.177, | |
| "eval_steps_per_second": 3.954, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0719671249389648, | |
| "eval_runtime": 0.498, | |
| "eval_samples_per_second": 72.294, | |
| "eval_steps_per_second": 4.016, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 5.022873401641846, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.3257, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0951011180877686, | |
| "eval_runtime": 0.4928, | |
| "eval_samples_per_second": 73.058, | |
| "eval_steps_per_second": 4.059, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 52.57142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0926969051361084, | |
| "eval_runtime": 0.503, | |
| "eval_samples_per_second": 71.576, | |
| "eval_steps_per_second": 3.976, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 53.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0882978439331055, | |
| "eval_runtime": 0.497, | |
| "eval_samples_per_second": 72.435, | |
| "eval_steps_per_second": 4.024, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0874247550964355, | |
| "eval_runtime": 0.4964, | |
| "eval_samples_per_second": 72.527, | |
| "eval_steps_per_second": 4.029, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.08829927444458, | |
| "eval_runtime": 0.4929, | |
| "eval_samples_per_second": 73.04, | |
| "eval_steps_per_second": 4.058, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 56.57142857142857, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0868247747421265, | |
| "eval_runtime": 0.5059, | |
| "eval_samples_per_second": 71.156, | |
| "eval_steps_per_second": 3.953, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 5.025148391723633, | |
| "learning_rate": 0.0, | |
| "loss": 0.321, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0857826471328735, | |
| "eval_runtime": 0.494, | |
| "eval_samples_per_second": 72.876, | |
| "eval_steps_per_second": 4.049, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "step": 100, | |
| "total_flos": 8.770779002957046e+17, | |
| "train_loss": 0.5723180460929871, | |
| "train_runtime": 553.2491, | |
| "train_samples_per_second": 35.789, | |
| "train_steps_per_second": 0.181 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.6944444444444444, | |
| "eval_loss": 0.9276576638221741, | |
| "eval_runtime": 0.5568, | |
| "eval_samples_per_second": 64.661, | |
| "eval_steps_per_second": 3.592, | |
| "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 | |
| } | |