| { | |
| "best_metric": 0.5555555555555556, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-hasta-65-fold5/checkpoint-59", | |
| "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": 0.3888888888888889, | |
| "eval_loss": 1.1751261949539185, | |
| "eval_runtime": 0.4879, | |
| "eval_samples_per_second": 73.787, | |
| "eval_steps_per_second": 4.099, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.0340763330459595, | |
| "eval_runtime": 0.489, | |
| "eval_samples_per_second": 73.621, | |
| "eval_steps_per_second": 4.09, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.2777777777777778, | |
| "eval_loss": 1.3058849573135376, | |
| "eval_runtime": 0.5086, | |
| "eval_samples_per_second": 70.788, | |
| "eval_steps_per_second": 3.933, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.2777777777777778, | |
| "eval_loss": 1.3255195617675781, | |
| "eval_runtime": 0.5171, | |
| "eval_samples_per_second": 69.615, | |
| "eval_steps_per_second": 3.868, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "eval_accuracy": 0.2777777777777778, | |
| "eval_loss": 1.1834439039230347, | |
| "eval_runtime": 0.4928, | |
| "eval_samples_per_second": 73.049, | |
| "eval_steps_per_second": 4.058, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 2.6557154655456543, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1839, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0356624126434326, | |
| "eval_runtime": 0.4894, | |
| "eval_samples_per_second": 73.563, | |
| "eval_steps_per_second": 4.087, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.060836672782898, | |
| "eval_runtime": 0.4874, | |
| "eval_samples_per_second": 73.855, | |
| "eval_steps_per_second": 4.103, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.3333333333333333, | |
| "eval_loss": 1.2059739828109741, | |
| "eval_runtime": 0.4931, | |
| "eval_samples_per_second": 73.015, | |
| "eval_steps_per_second": 4.056, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.1938085556030273, | |
| "eval_runtime": 0.4908, | |
| "eval_samples_per_second": 73.347, | |
| "eval_steps_per_second": 4.075, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.082518219947815, | |
| "eval_runtime": 0.4851, | |
| "eval_samples_per_second": 74.208, | |
| "eval_steps_per_second": 4.123, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.148760199546814, | |
| "eval_runtime": 0.4962, | |
| "eval_samples_per_second": 72.544, | |
| "eval_steps_per_second": 4.03, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 3.6616809368133545, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.9707, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.1268404722213745, | |
| "eval_runtime": 0.4917, | |
| "eval_samples_per_second": 73.22, | |
| "eval_steps_per_second": 4.068, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0563452243804932, | |
| "eval_runtime": 0.4999, | |
| "eval_samples_per_second": 72.018, | |
| "eval_steps_per_second": 4.001, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0570094585418701, | |
| "eval_runtime": 0.4934, | |
| "eval_samples_per_second": 72.957, | |
| "eval_steps_per_second": 4.053, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.1166433095932007, | |
| "eval_runtime": 0.4883, | |
| "eval_samples_per_second": 73.728, | |
| "eval_steps_per_second": 4.096, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.0609068870544434, | |
| "eval_runtime": 0.4936, | |
| "eval_samples_per_second": 72.941, | |
| "eval_steps_per_second": 4.052, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 16.571428571428573, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.0378516912460327, | |
| "eval_runtime": 0.4863, | |
| "eval_samples_per_second": 74.034, | |
| "eval_steps_per_second": 4.113, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 4.919348239898682, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.8668, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 17.714285714285715, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.0610023736953735, | |
| "eval_runtime": 0.4887, | |
| "eval_samples_per_second": 73.661, | |
| "eval_steps_per_second": 4.092, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.1810537576675415, | |
| "eval_runtime": 0.4883, | |
| "eval_samples_per_second": 73.727, | |
| "eval_steps_per_second": 4.096, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.102830410003662, | |
| "eval_runtime": 0.5057, | |
| "eval_samples_per_second": 71.192, | |
| "eval_steps_per_second": 3.955, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 20.571428571428573, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.0949718952178955, | |
| "eval_runtime": 0.4917, | |
| "eval_samples_per_second": 73.219, | |
| "eval_steps_per_second": 4.068, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 21.714285714285715, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.1424427032470703, | |
| "eval_runtime": 0.4936, | |
| "eval_samples_per_second": 72.934, | |
| "eval_steps_per_second": 4.052, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 6.751219272613525, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6889, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.3027087450027466, | |
| "eval_runtime": 0.489, | |
| "eval_samples_per_second": 73.623, | |
| "eval_steps_per_second": 4.09, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.2030115127563477, | |
| "eval_runtime": 0.4898, | |
| "eval_samples_per_second": 73.499, | |
| "eval_steps_per_second": 4.083, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 24.571428571428573, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.2148040533065796, | |
| "eval_runtime": 0.5046, | |
| "eval_samples_per_second": 71.339, | |
| "eval_steps_per_second": 3.963, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.3066232204437256, | |
| "eval_runtime": 0.4837, | |
| "eval_samples_per_second": 74.426, | |
| "eval_steps_per_second": 4.135, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.3611111111111111, | |
| "eval_loss": 1.388055443763733, | |
| "eval_runtime": 0.4929, | |
| "eval_samples_per_second": 73.037, | |
| "eval_steps_per_second": 4.058, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.2565866708755493, | |
| "eval_runtime": 0.4954, | |
| "eval_samples_per_second": 72.666, | |
| "eval_steps_per_second": 4.037, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 4.931034564971924, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.576, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.1890705823898315, | |
| "eval_runtime": 0.485, | |
| "eval_samples_per_second": 74.228, | |
| "eval_steps_per_second": 4.124, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 29.714285714285715, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.163759708404541, | |
| "eval_runtime": 0.4917, | |
| "eval_samples_per_second": 73.21, | |
| "eval_steps_per_second": 4.067, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.253006935119629, | |
| "eval_runtime": 0.4917, | |
| "eval_samples_per_second": 73.209, | |
| "eval_steps_per_second": 4.067, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.13831627368927, | |
| "eval_runtime": 0.4952, | |
| "eval_samples_per_second": 72.692, | |
| "eval_steps_per_second": 4.038, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 32.57142857142857, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0968199968338013, | |
| "eval_runtime": 0.4993, | |
| "eval_samples_per_second": 72.096, | |
| "eval_steps_per_second": 4.005, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 33.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0162580013275146, | |
| "eval_runtime": 0.4924, | |
| "eval_samples_per_second": 73.107, | |
| "eval_steps_per_second": 4.061, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 6.734873294830322, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.4773, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.1107090711593628, | |
| "eval_runtime": 0.4902, | |
| "eval_samples_per_second": 73.44, | |
| "eval_steps_per_second": 4.08, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.1340625286102295, | |
| "eval_runtime": 0.4891, | |
| "eval_samples_per_second": 73.602, | |
| "eval_steps_per_second": 4.089, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 36.57142857142857, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.1151530742645264, | |
| "eval_runtime": 0.4831, | |
| "eval_samples_per_second": 74.516, | |
| "eval_steps_per_second": 4.14, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 37.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.115782380104065, | |
| "eval_runtime": 0.5062, | |
| "eval_samples_per_second": 71.115, | |
| "eval_steps_per_second": 3.951, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.1627553701400757, | |
| "eval_runtime": 0.4965, | |
| "eval_samples_per_second": 72.512, | |
| "eval_steps_per_second": 4.028, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 4.625010967254639, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.4186, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.2304986715316772, | |
| "eval_runtime": 0.5035, | |
| "eval_samples_per_second": 71.502, | |
| "eval_steps_per_second": 3.972, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.57142857142857, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.2181116342544556, | |
| "eval_runtime": 0.4887, | |
| "eval_samples_per_second": 73.665, | |
| "eval_steps_per_second": 4.093, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 41.714285714285715, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.216411828994751, | |
| "eval_runtime": 0.4978, | |
| "eval_samples_per_second": 72.325, | |
| "eval_steps_per_second": 4.018, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.2224550247192383, | |
| "eval_runtime": 0.5007, | |
| "eval_samples_per_second": 71.898, | |
| "eval_steps_per_second": 3.994, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.2297773361206055, | |
| "eval_runtime": 0.492, | |
| "eval_samples_per_second": 73.173, | |
| "eval_steps_per_second": 4.065, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 44.57142857142857, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.2651033401489258, | |
| "eval_runtime": 0.4869, | |
| "eval_samples_per_second": 73.937, | |
| "eval_steps_per_second": 4.108, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 4.358370304107666, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.3318, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.3628058433532715, | |
| "eval_runtime": 0.4972, | |
| "eval_samples_per_second": 72.408, | |
| "eval_steps_per_second": 4.023, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.3817496299743652, | |
| "eval_runtime": 0.5101, | |
| "eval_samples_per_second": 70.572, | |
| "eval_steps_per_second": 3.921, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.359439492225647, | |
| "eval_runtime": 0.5012, | |
| "eval_samples_per_second": 71.825, | |
| "eval_steps_per_second": 3.99, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.355283260345459, | |
| "eval_runtime": 0.4956, | |
| "eval_samples_per_second": 72.644, | |
| "eval_steps_per_second": 4.036, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 49.714285714285715, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.3547987937927246, | |
| "eval_runtime": 0.4823, | |
| "eval_samples_per_second": 74.639, | |
| "eval_steps_per_second": 4.147, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.4113057851791382, | |
| "eval_runtime": 0.4928, | |
| "eval_samples_per_second": 73.049, | |
| "eval_steps_per_second": 4.058, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 5.402642726898193, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.344, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.4433233737945557, | |
| "eval_runtime": 0.4954, | |
| "eval_samples_per_second": 72.67, | |
| "eval_steps_per_second": 4.037, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 52.57142857142857, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.4449433088302612, | |
| "eval_runtime": 0.5015, | |
| "eval_samples_per_second": 71.781, | |
| "eval_steps_per_second": 3.988, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 53.714285714285715, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.4513975381851196, | |
| "eval_runtime": 0.4877, | |
| "eval_samples_per_second": 73.809, | |
| "eval_steps_per_second": 4.101, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.4684746265411377, | |
| "eval_runtime": 0.4939, | |
| "eval_samples_per_second": 72.889, | |
| "eval_steps_per_second": 4.049, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.473361611366272, | |
| "eval_runtime": 0.5076, | |
| "eval_samples_per_second": 70.919, | |
| "eval_steps_per_second": 3.94, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 56.57142857142857, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.4746534824371338, | |
| "eval_runtime": 0.4963, | |
| "eval_samples_per_second": 72.541, | |
| "eval_steps_per_second": 4.03, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 5.429214000701904, | |
| "learning_rate": 0.0, | |
| "loss": 0.3305, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.4732346534729004, | |
| "eval_runtime": 0.4898, | |
| "eval_samples_per_second": 73.493, | |
| "eval_steps_per_second": 4.083, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "step": 100, | |
| "total_flos": 8.903293865079828e+17, | |
| "train_loss": 0.6188502907752991, | |
| "train_runtime": 552.0025, | |
| "train_samples_per_second": 36.413, | |
| "train_steps_per_second": 0.181 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0162580013275146, | |
| "eval_runtime": 0.5216, | |
| "eval_samples_per_second": 69.013, | |
| "eval_steps_per_second": 3.834, | |
| "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 | |
| } | |