| { | |
| "best_metric": 0.6111111111111112, | |
| "best_model_checkpoint": "beit-base-patch16-224-hasta-65-fold3/checkpoint-68", | |
| "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": 1.2972347736358643, | |
| "eval_runtime": 0.4909, | |
| "eval_samples_per_second": 73.334, | |
| "eval_steps_per_second": 4.074, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.1961228847503662, | |
| "eval_runtime": 0.4989, | |
| "eval_samples_per_second": 72.158, | |
| "eval_steps_per_second": 4.009, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.2777777777777778, | |
| "eval_loss": 1.5407676696777344, | |
| "eval_runtime": 0.5093, | |
| "eval_samples_per_second": 70.691, | |
| "eval_steps_per_second": 3.927, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3611111111111111, | |
| "eval_loss": 1.1727068424224854, | |
| "eval_runtime": 0.5013, | |
| "eval_samples_per_second": 71.816, | |
| "eval_steps_per_second": 3.99, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "eval_accuracy": 0.3611111111111111, | |
| "eval_loss": 1.141831874847412, | |
| "eval_runtime": 0.4972, | |
| "eval_samples_per_second": 72.411, | |
| "eval_steps_per_second": 4.023, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 5.564121723175049, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2077, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "eval_accuracy": 0.2777777777777778, | |
| "eval_loss": 1.2295489311218262, | |
| "eval_runtime": 0.4887, | |
| "eval_samples_per_second": 73.67, | |
| "eval_steps_per_second": 4.093, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.1357738971710205, | |
| "eval_runtime": 0.5087, | |
| "eval_samples_per_second": 70.766, | |
| "eval_steps_per_second": 3.931, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 1.139593482017517, | |
| "eval_runtime": 0.4968, | |
| "eval_samples_per_second": 72.463, | |
| "eval_steps_per_second": 4.026, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_accuracy": 0.3611111111111111, | |
| "eval_loss": 1.1826934814453125, | |
| "eval_runtime": 0.4968, | |
| "eval_samples_per_second": 72.47, | |
| "eval_steps_per_second": 4.026, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "eval_accuracy": 0.3888888888888889, | |
| "eval_loss": 1.071872591972351, | |
| "eval_runtime": 0.5161, | |
| "eval_samples_per_second": 69.752, | |
| "eval_steps_per_second": 3.875, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.4166666666666667, | |
| "eval_loss": 1.0910537242889404, | |
| "eval_runtime": 0.5152, | |
| "eval_samples_per_second": 69.88, | |
| "eval_steps_per_second": 3.882, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 6.144622325897217, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 1.004, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0003440380096436, | |
| "eval_runtime": 0.4949, | |
| "eval_samples_per_second": 72.737, | |
| "eval_steps_per_second": 4.041, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.9456803202629089, | |
| "eval_runtime": 0.4946, | |
| "eval_samples_per_second": 72.793, | |
| "eval_steps_per_second": 4.044, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0181926488876343, | |
| "eval_runtime": 0.5019, | |
| "eval_samples_per_second": 71.726, | |
| "eval_steps_per_second": 3.985, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.9877431392669678, | |
| "eval_runtime": 0.4912, | |
| "eval_samples_per_second": 73.291, | |
| "eval_steps_per_second": 4.072, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9536970853805542, | |
| "eval_runtime": 0.4946, | |
| "eval_samples_per_second": 72.783, | |
| "eval_steps_per_second": 4.043, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 16.571428571428573, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9310481548309326, | |
| "eval_runtime": 0.5056, | |
| "eval_samples_per_second": 71.199, | |
| "eval_steps_per_second": 3.955, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 8.827644348144531, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.8983, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 17.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.8763439059257507, | |
| "eval_runtime": 0.5026, | |
| "eval_samples_per_second": 71.621, | |
| "eval_steps_per_second": 3.979, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.9626156687736511, | |
| "eval_runtime": 0.4971, | |
| "eval_samples_per_second": 72.421, | |
| "eval_steps_per_second": 4.023, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0251946449279785, | |
| "eval_runtime": 0.5127, | |
| "eval_samples_per_second": 70.213, | |
| "eval_steps_per_second": 3.901, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 20.571428571428573, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0288382768630981, | |
| "eval_runtime": 0.4975, | |
| "eval_samples_per_second": 72.364, | |
| "eval_steps_per_second": 4.02, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 21.714285714285715, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.0807000398635864, | |
| "eval_runtime": 0.4959, | |
| "eval_samples_per_second": 72.592, | |
| "eval_steps_per_second": 4.033, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 7.877163887023926, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.7347, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.4444444444444444, | |
| "eval_loss": 0.9994614124298096, | |
| "eval_runtime": 0.5076, | |
| "eval_samples_per_second": 70.923, | |
| "eval_steps_per_second": 3.94, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.3380463123321533, | |
| "eval_runtime": 0.4927, | |
| "eval_samples_per_second": 73.066, | |
| "eval_steps_per_second": 4.059, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 24.571428571428573, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.3829317092895508, | |
| "eval_runtime": 0.4915, | |
| "eval_samples_per_second": 73.25, | |
| "eval_steps_per_second": 4.069, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "eval_accuracy": 0.4722222222222222, | |
| "eval_loss": 1.0370383262634277, | |
| "eval_runtime": 0.4924, | |
| "eval_samples_per_second": 73.115, | |
| "eval_steps_per_second": 4.062, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.037151575088501, | |
| "eval_runtime": 0.4905, | |
| "eval_samples_per_second": 73.398, | |
| "eval_steps_per_second": 4.078, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9477570056915283, | |
| "eval_runtime": 0.4986, | |
| "eval_samples_per_second": 72.202, | |
| "eval_steps_per_second": 4.011, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 6.354820728302002, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.6583, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9327968955039978, | |
| "eval_runtime": 0.4991, | |
| "eval_samples_per_second": 72.136, | |
| "eval_steps_per_second": 4.008, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 29.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 1.0591890811920166, | |
| "eval_runtime": 0.4917, | |
| "eval_samples_per_second": 73.217, | |
| "eval_steps_per_second": 4.068, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9613094329833984, | |
| "eval_runtime": 0.4952, | |
| "eval_samples_per_second": 72.691, | |
| "eval_steps_per_second": 4.038, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 0.8314970135688782, | |
| "eval_runtime": 0.4969, | |
| "eval_samples_per_second": 72.443, | |
| "eval_steps_per_second": 4.025, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 32.57142857142857, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.8748820424079895, | |
| "eval_runtime": 0.4978, | |
| "eval_samples_per_second": 72.314, | |
| "eval_steps_per_second": 4.017, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 33.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.038988471031189, | |
| "eval_runtime": 0.5018, | |
| "eval_samples_per_second": 71.743, | |
| "eval_steps_per_second": 3.986, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 8.402383804321289, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.5633, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9446022510528564, | |
| "eval_runtime": 0.4971, | |
| "eval_samples_per_second": 72.424, | |
| "eval_steps_per_second": 4.024, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.958649218082428, | |
| "eval_runtime": 0.4937, | |
| "eval_samples_per_second": 72.924, | |
| "eval_steps_per_second": 4.051, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 36.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.987652063369751, | |
| "eval_runtime": 0.4944, | |
| "eval_samples_per_second": 72.812, | |
| "eval_steps_per_second": 4.045, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 37.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0394772291183472, | |
| "eval_runtime": 0.497, | |
| "eval_samples_per_second": 72.441, | |
| "eval_steps_per_second": 4.024, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9493582844734192, | |
| "eval_runtime": 0.5087, | |
| "eval_samples_per_second": 70.774, | |
| "eval_steps_per_second": 3.932, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 10.07642650604248, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.4905, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.012374758720398, | |
| "eval_runtime": 0.4923, | |
| "eval_samples_per_second": 73.122, | |
| "eval_steps_per_second": 4.062, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.57142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.1536669731140137, | |
| "eval_runtime": 0.5085, | |
| "eval_samples_per_second": 70.8, | |
| "eval_steps_per_second": 3.933, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 41.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.2837426662445068, | |
| "eval_runtime": 0.4943, | |
| "eval_samples_per_second": 72.823, | |
| "eval_steps_per_second": 4.046, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.082094430923462, | |
| "eval_runtime": 0.4953, | |
| "eval_samples_per_second": 72.686, | |
| "eval_steps_per_second": 4.038, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.0084452629089355, | |
| "eval_runtime": 0.4992, | |
| "eval_samples_per_second": 72.117, | |
| "eval_steps_per_second": 4.006, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 44.57142857142857, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.043327808380127, | |
| "eval_runtime": 0.5076, | |
| "eval_samples_per_second": 70.922, | |
| "eval_steps_per_second": 3.94, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 5.775242805480957, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.4609, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.211928367614746, | |
| "eval_runtime": 0.4968, | |
| "eval_samples_per_second": 72.468, | |
| "eval_steps_per_second": 4.026, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.5555555555555556, | |
| "eval_loss": 1.2975627183914185, | |
| "eval_runtime": 0.5093, | |
| "eval_samples_per_second": 70.69, | |
| "eval_steps_per_second": 3.927, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.14437997341156, | |
| "eval_runtime": 0.4953, | |
| "eval_samples_per_second": 72.687, | |
| "eval_steps_per_second": 4.038, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.04435133934021, | |
| "eval_runtime": 0.497, | |
| "eval_samples_per_second": 72.439, | |
| "eval_steps_per_second": 4.024, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 49.714285714285715, | |
| "eval_accuracy": 0.5277777777777778, | |
| "eval_loss": 0.9337983727455139, | |
| "eval_runtime": 0.505, | |
| "eval_samples_per_second": 71.292, | |
| "eval_steps_per_second": 3.961, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9500620365142822, | |
| "eval_runtime": 0.5149, | |
| "eval_samples_per_second": 69.912, | |
| "eval_steps_per_second": 3.884, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 10.107069969177246, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.4106, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0080580711364746, | |
| "eval_runtime": 0.5027, | |
| "eval_samples_per_second": 71.615, | |
| "eval_steps_per_second": 3.979, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 52.57142857142857, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0579811334609985, | |
| "eval_runtime": 0.507, | |
| "eval_samples_per_second": 71.001, | |
| "eval_steps_per_second": 3.944, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 53.714285714285715, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.1187769174575806, | |
| "eval_runtime": 0.4951, | |
| "eval_samples_per_second": 72.717, | |
| "eval_steps_per_second": 4.04, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.129145622253418, | |
| "eval_runtime": 0.5123, | |
| "eval_samples_per_second": 70.274, | |
| "eval_steps_per_second": 3.904, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.1155930757522583, | |
| "eval_runtime": 0.502, | |
| "eval_samples_per_second": 71.707, | |
| "eval_steps_per_second": 3.984, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 56.57142857142857, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.1098636388778687, | |
| "eval_runtime": 0.5016, | |
| "eval_samples_per_second": 71.776, | |
| "eval_steps_per_second": 3.988, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 5.268594741821289, | |
| "learning_rate": 0.0, | |
| "loss": 0.3784, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 1.1064138412475586, | |
| "eval_runtime": 0.5198, | |
| "eval_samples_per_second": 69.256, | |
| "eval_steps_per_second": 3.848, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "step": 100, | |
| "total_flos": 8.899329213047992e+17, | |
| "train_loss": 0.6806672286987304, | |
| "train_runtime": 554.0122, | |
| "train_samples_per_second": 36.281, | |
| "train_steps_per_second": 0.181 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "eval_accuracy": 0.6111111111111112, | |
| "eval_loss": 0.9493582844734192, | |
| "eval_runtime": 0.5666, | |
| "eval_samples_per_second": 63.533, | |
| "eval_steps_per_second": 3.53, | |
| "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 | |
| } | |