| { | |
| "best_metric": 0.8225806451612904, | |
| "best_model_checkpoint": "beit-base-patch16-224-OT\\checkpoint-99", | |
| "epoch": 35.55555555555556, | |
| "eval_steps": 500, | |
| "global_step": 160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.89, | |
| "eval_accuracy": 0.14516129032258066, | |
| "eval_loss": 1.7602994441986084, | |
| "eval_runtime": 2.148, | |
| "eval_samples_per_second": 28.864, | |
| "eval_steps_per_second": 1.862, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.14516129032258066, | |
| "eval_loss": 1.685233473777771, | |
| "eval_runtime": 2.219, | |
| "eval_samples_per_second": 27.94, | |
| "eval_steps_per_second": 1.803, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.7571, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "eval_accuracy": 0.14516129032258066, | |
| "eval_loss": 1.5655227899551392, | |
| "eval_runtime": 2.222, | |
| "eval_samples_per_second": 27.903, | |
| "eval_steps_per_second": 1.8, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.14516129032258066, | |
| "eval_loss": 1.3815577030181885, | |
| "eval_runtime": 2.3696, | |
| "eval_samples_per_second": 26.165, | |
| "eval_steps_per_second": 1.688, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 1.5255, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "eval_accuracy": 0.3225806451612903, | |
| "eval_loss": 1.2598901987075806, | |
| "eval_runtime": 2.178, | |
| "eval_samples_per_second": 28.466, | |
| "eval_steps_per_second": 1.837, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.4838709677419355, | |
| "eval_loss": 1.153410792350769, | |
| "eval_runtime": 2.227, | |
| "eval_samples_per_second": 27.84, | |
| "eval_steps_per_second": 1.796, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 1.2245, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "eval_accuracy": 0.4838709677419355, | |
| "eval_loss": 1.0641188621520996, | |
| "eval_runtime": 2.203, | |
| "eval_samples_per_second": 28.143, | |
| "eval_steps_per_second": 1.816, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.43548387096774194, | |
| "eval_loss": 1.0371758937835693, | |
| "eval_runtime": 2.2355, | |
| "eval_samples_per_second": 27.734, | |
| "eval_steps_per_second": 1.789, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.0438, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "eval_accuracy": 0.43548387096774194, | |
| "eval_loss": 0.9987961649894714, | |
| "eval_runtime": 2.3086, | |
| "eval_samples_per_second": 26.857, | |
| "eval_steps_per_second": 1.733, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5161290322580645, | |
| "eval_loss": 0.9259945154190063, | |
| "eval_runtime": 2.2905, | |
| "eval_samples_per_second": 27.068, | |
| "eval_steps_per_second": 1.746, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 10.89, | |
| "eval_accuracy": 0.7096774193548387, | |
| "eval_loss": 0.9084866046905518, | |
| "eval_runtime": 2.2545, | |
| "eval_samples_per_second": 27.5, | |
| "eval_steps_per_second": 1.774, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 0.9727, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7258064516129032, | |
| "eval_loss": 0.843325674533844, | |
| "eval_runtime": 2.1865, | |
| "eval_samples_per_second": 28.355, | |
| "eval_steps_per_second": 1.829, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 12.89, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.7529016137123108, | |
| "eval_runtime": 2.1995, | |
| "eval_samples_per_second": 28.188, | |
| "eval_steps_per_second": 1.819, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.8469, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.7186572551727295, | |
| "eval_runtime": 2.2225, | |
| "eval_samples_per_second": 27.896, | |
| "eval_steps_per_second": 1.8, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 14.89, | |
| "eval_accuracy": 0.7258064516129032, | |
| "eval_loss": 0.6805762648582458, | |
| "eval_runtime": 2.1855, | |
| "eval_samples_per_second": 28.369, | |
| "eval_steps_per_second": 1.83, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 15.56, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.6908, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.6575707197189331, | |
| "eval_runtime": 2.2315, | |
| "eval_samples_per_second": 27.784, | |
| "eval_steps_per_second": 1.792, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "eval_accuracy": 0.7903225806451613, | |
| "eval_loss": 0.5742202997207642, | |
| "eval_runtime": 2.261, | |
| "eval_samples_per_second": 27.421, | |
| "eval_steps_per_second": 1.769, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 17.78, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.6064, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.6446634531021118, | |
| "eval_runtime": 2.3311, | |
| "eval_samples_per_second": 26.597, | |
| "eval_steps_per_second": 1.716, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 18.89, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.5602142810821533, | |
| "eval_runtime": 2.1895, | |
| "eval_samples_per_second": 28.317, | |
| "eval_steps_per_second": 1.827, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 0.5303, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7903225806451613, | |
| "eval_loss": 0.49427932500839233, | |
| "eval_runtime": 2.2745, | |
| "eval_samples_per_second": 27.258, | |
| "eval_steps_per_second": 1.759, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 20.89, | |
| "eval_accuracy": 0.7903225806451613, | |
| "eval_loss": 0.530381441116333, | |
| "eval_runtime": 2.142, | |
| "eval_samples_per_second": 28.945, | |
| "eval_steps_per_second": 1.867, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.48010584712028503, | |
| "eval_runtime": 2.3616, | |
| "eval_samples_per_second": 26.254, | |
| "eval_steps_per_second": 1.694, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.4903, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 22.89, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.4848884344100952, | |
| "eval_runtime": 2.16, | |
| "eval_samples_per_second": 28.703, | |
| "eval_steps_per_second": 1.852, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.5709980726242065, | |
| "eval_runtime": 2.1685, | |
| "eval_samples_per_second": 28.591, | |
| "eval_steps_per_second": 1.845, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 24.44, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 0.4261, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 24.89, | |
| "eval_accuracy": 0.7903225806451613, | |
| "eval_loss": 0.4803168475627899, | |
| "eval_runtime": 2.2145, | |
| "eval_samples_per_second": 27.997, | |
| "eval_steps_per_second": 1.806, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.7258064516129032, | |
| "eval_loss": 0.5670634508132935, | |
| "eval_runtime": 2.1993, | |
| "eval_samples_per_second": 28.191, | |
| "eval_steps_per_second": 1.819, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.4122, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 26.89, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.4585064947605133, | |
| "eval_runtime": 2.2065, | |
| "eval_samples_per_second": 28.098, | |
| "eval_steps_per_second": 1.813, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7096774193548387, | |
| "eval_loss": 0.5910329222679138, | |
| "eval_runtime": 2.23, | |
| "eval_samples_per_second": 27.802, | |
| "eval_steps_per_second": 1.794, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 28.89, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 0.3739, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 28.89, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.5821260213851929, | |
| "eval_runtime": 2.2145, | |
| "eval_samples_per_second": 27.997, | |
| "eval_steps_per_second": 1.806, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.5329306125640869, | |
| "eval_runtime": 2.3526, | |
| "eval_samples_per_second": 26.354, | |
| "eval_steps_per_second": 1.7, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 30.89, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.4422537386417389, | |
| "eval_runtime": 2.1625, | |
| "eval_samples_per_second": 28.67, | |
| "eval_steps_per_second": 1.85, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 31.11, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.3896, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.47155243158340454, | |
| "eval_runtime": 2.1865, | |
| "eval_samples_per_second": 28.356, | |
| "eval_steps_per_second": 1.829, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 32.89, | |
| "eval_accuracy": 0.7903225806451613, | |
| "eval_loss": 0.4785827100276947, | |
| "eval_runtime": 2.232, | |
| "eval_samples_per_second": 27.777, | |
| "eval_steps_per_second": 1.792, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 0.3472, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.7903225806451613, | |
| "eval_loss": 0.45382845401763916, | |
| "eval_runtime": 2.19, | |
| "eval_samples_per_second": 28.31, | |
| "eval_steps_per_second": 1.826, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 34.89, | |
| "eval_accuracy": 0.7903225806451613, | |
| "eval_loss": 0.455282062292099, | |
| "eval_runtime": 2.2165, | |
| "eval_samples_per_second": 27.972, | |
| "eval_steps_per_second": 1.805, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 35.56, | |
| "learning_rate": 0.0, | |
| "loss": 0.3349, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 35.56, | |
| "eval_accuracy": 0.7903225806451613, | |
| "eval_loss": 0.45279815793037415, | |
| "eval_runtime": 2.4931, | |
| "eval_samples_per_second": 24.869, | |
| "eval_steps_per_second": 1.604, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 35.56, | |
| "step": 160, | |
| "total_flos": 7.931930389512192e+17, | |
| "train_loss": 0.7482577681541442, | |
| "train_runtime": 589.7762, | |
| "train_samples_per_second": 19.533, | |
| "train_steps_per_second": 0.271 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 160, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 7.931930389512192e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |