| { |
| "best_metric": 0.0012013022787868977, |
| "best_model_checkpoint": "./vit-huge-modified-augmented-ph2-patch-14/checkpoint-650", |
| "epoch": 4.0, |
| "global_step": 680, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.06, |
| "learning_rate": 0.00019705882352941177, |
| "loss": 0.4297, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.12, |
| "learning_rate": 0.00019411764705882354, |
| "loss": 0.2305, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.18, |
| "learning_rate": 0.0001911764705882353, |
| "loss": 0.191, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.24, |
| "learning_rate": 0.00018823529411764707, |
| "loss": 0.0725, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.29, |
| "learning_rate": 0.00018529411764705883, |
| "loss": 0.0996, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.29, |
| "eval_accuracy": 0.9366438356164384, |
| "eval_f1": 0.9366438356164383, |
| "eval_loss": 0.137769415974617, |
| "eval_precision": 0.9366438356164384, |
| "eval_recall": 0.9366438356164384, |
| "eval_runtime": 12.0421, |
| "eval_samples_per_second": 48.497, |
| "eval_steps_per_second": 6.062, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.35, |
| "learning_rate": 0.0001823529411764706, |
| "loss": 0.0622, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.41, |
| "learning_rate": 0.00017941176470588236, |
| "loss": 0.0802, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.47, |
| "learning_rate": 0.00017647058823529413, |
| "loss": 0.0313, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.53, |
| "learning_rate": 0.0001735294117647059, |
| "loss": 0.0278, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.59, |
| "learning_rate": 0.00017058823529411766, |
| "loss": 0.0096, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.59, |
| "eval_accuracy": 0.9743150684931506, |
| "eval_f1": 0.9743150684931506, |
| "eval_loss": 0.05090475082397461, |
| "eval_precision": 0.9743150684931506, |
| "eval_recall": 0.9743150684931506, |
| "eval_runtime": 12.3686, |
| "eval_samples_per_second": 47.217, |
| "eval_steps_per_second": 5.902, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.65, |
| "learning_rate": 0.00016764705882352942, |
| "loss": 0.0087, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.71, |
| "learning_rate": 0.0001647058823529412, |
| "loss": 0.0274, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.76, |
| "learning_rate": 0.00016176470588235295, |
| "loss": 0.0053, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.82, |
| "learning_rate": 0.0001588235294117647, |
| "loss": 0.0048, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.88, |
| "learning_rate": 0.00015588235294117648, |
| "loss": 0.0049, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.88, |
| "eval_accuracy": 0.9982876712328768, |
| "eval_f1": 0.9982876712328768, |
| "eval_loss": 0.008507030084729195, |
| "eval_precision": 0.9982876712328768, |
| "eval_recall": 0.9982876712328768, |
| "eval_runtime": 12.0996, |
| "eval_samples_per_second": 48.266, |
| "eval_steps_per_second": 6.033, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.94, |
| "learning_rate": 0.00015294117647058822, |
| "loss": 0.0045, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.0, |
| "learning_rate": 0.00015000000000000001, |
| "loss": 0.0037, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.06, |
| "learning_rate": 0.00014705882352941178, |
| "loss": 0.0034, |
| "step": 180 |
| }, |
| { |
| "epoch": 1.12, |
| "learning_rate": 0.00014411764705882354, |
| "loss": 0.0032, |
| "step": 190 |
| }, |
| { |
| "epoch": 1.18, |
| "learning_rate": 0.0001411764705882353, |
| "loss": 0.0029, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.18, |
| "eval_accuracy": 1.0, |
| "eval_f1": 1.0, |
| "eval_loss": 0.00365791586227715, |
| "eval_precision": 1.0, |
| "eval_recall": 1.0, |
| "eval_runtime": 12.0585, |
| "eval_samples_per_second": 48.431, |
| "eval_steps_per_second": 6.054, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.24, |
| "learning_rate": 0.00013823529411764707, |
| "loss": 0.0028, |
| "step": 210 |
| }, |
| { |
| "epoch": 1.29, |
| "learning_rate": 0.00013529411764705884, |
| "loss": 0.0026, |
| "step": 220 |
| }, |
| { |
| "epoch": 1.35, |
| "learning_rate": 0.0001323529411764706, |
| "loss": 0.0024, |
| "step": 230 |
| }, |
| { |
| "epoch": 1.41, |
| "learning_rate": 0.00012941176470588237, |
| "loss": 0.0023, |
| "step": 240 |
| }, |
| { |
| "epoch": 1.47, |
| "learning_rate": 0.0001264705882352941, |
| "loss": 0.0022, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.47, |
| "eval_accuracy": 1.0, |
| "eval_f1": 1.0, |
| "eval_loss": 0.002802133560180664, |
| "eval_precision": 1.0, |
| "eval_recall": 1.0, |
| "eval_runtime": 12.043, |
| "eval_samples_per_second": 48.493, |
| "eval_steps_per_second": 6.062, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.53, |
| "learning_rate": 0.0001235294117647059, |
| "loss": 0.0022, |
| "step": 260 |
| }, |
| { |
| "epoch": 1.59, |
| "learning_rate": 0.00012058823529411765, |
| "loss": 0.002, |
| "step": 270 |
| }, |
| { |
| "epoch": 1.65, |
| "learning_rate": 0.00011764705882352942, |
| "loss": 0.002, |
| "step": 280 |
| }, |
| { |
| "epoch": 1.71, |
| "learning_rate": 0.00011470588235294118, |
| "loss": 0.0018, |
| "step": 290 |
| }, |
| { |
| "epoch": 1.76, |
| "learning_rate": 0.00011176470588235294, |
| "loss": 0.0018, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_accuracy": 1.0, |
| "eval_f1": 1.0, |
| "eval_loss": 0.002224571071565151, |
| "eval_precision": 1.0, |
| "eval_recall": 1.0, |
| "eval_runtime": 12.0863, |
| "eval_samples_per_second": 48.319, |
| "eval_steps_per_second": 6.04, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.82, |
| "learning_rate": 0.0001088235294117647, |
| "loss": 0.0017, |
| "step": 310 |
| }, |
| { |
| "epoch": 1.88, |
| "learning_rate": 0.00010588235294117647, |
| "loss": 0.0017, |
| "step": 320 |
| }, |
| { |
| "epoch": 1.94, |
| "learning_rate": 0.00010294117647058823, |
| "loss": 0.0016, |
| "step": 330 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 0.0001, |
| "loss": 0.0015, |
| "step": 340 |
| }, |
| { |
| "epoch": 2.06, |
| "learning_rate": 9.705882352941177e-05, |
| "loss": 0.0015, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.06, |
| "eval_accuracy": 1.0, |
| "eval_f1": 1.0, |
| "eval_loss": 0.0020704269409179688, |
| "eval_precision": 1.0, |
| "eval_recall": 1.0, |
| "eval_runtime": 12.1137, |
| "eval_samples_per_second": 48.21, |
| "eval_steps_per_second": 6.026, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.12, |
| "learning_rate": 9.411764705882353e-05, |
| "loss": 0.0015, |
| "step": 360 |
| }, |
| { |
| "epoch": 2.18, |
| "learning_rate": 9.11764705882353e-05, |
| "loss": 0.0014, |
| "step": 370 |
| }, |
| { |
| "epoch": 2.24, |
| "learning_rate": 8.823529411764706e-05, |
| "loss": 0.0014, |
| "step": 380 |
| }, |
| { |
| "epoch": 2.29, |
| "learning_rate": 8.529411764705883e-05, |
| "loss": 0.0013, |
| "step": 390 |
| }, |
| { |
| "epoch": 2.35, |
| "learning_rate": 8.23529411764706e-05, |
| "loss": 0.0013, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.35, |
| "eval_accuracy": 1.0, |
| "eval_f1": 1.0, |
| "eval_loss": 0.001730100717395544, |
| "eval_precision": 1.0, |
| "eval_recall": 1.0, |
| "eval_runtime": 12.202, |
| "eval_samples_per_second": 47.861, |
| "eval_steps_per_second": 5.983, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.41, |
| "learning_rate": 7.941176470588235e-05, |
| "loss": 0.0013, |
| "step": 410 |
| }, |
| { |
| "epoch": 2.47, |
| "learning_rate": 7.647058823529411e-05, |
| "loss": 0.0012, |
| "step": 420 |
| }, |
| { |
| "epoch": 2.53, |
| "learning_rate": 7.352941176470589e-05, |
| "loss": 0.0012, |
| "step": 430 |
| }, |
| { |
| "epoch": 2.59, |
| "learning_rate": 7.058823529411765e-05, |
| "loss": 0.0012, |
| "step": 440 |
| }, |
| { |
| "epoch": 2.65, |
| "learning_rate": 6.764705882352942e-05, |
| "loss": 0.0011, |
| "step": 450 |
| }, |
| { |
| "epoch": 2.65, |
| "eval_accuracy": 1.0, |
| "eval_f1": 1.0, |
| "eval_loss": 0.0015468703350052238, |
| "eval_precision": 1.0, |
| "eval_recall": 1.0, |
| "eval_runtime": 11.9628, |
| "eval_samples_per_second": 48.818, |
| "eval_steps_per_second": 6.102, |
| "step": 450 |
| }, |
| { |
| "epoch": 2.71, |
| "learning_rate": 6.470588235294118e-05, |
| "loss": 0.0011, |
| "step": 460 |
| }, |
| { |
| "epoch": 2.76, |
| "learning_rate": 6.176470588235295e-05, |
| "loss": 0.0012, |
| "step": 470 |
| }, |
| { |
| "epoch": 2.82, |
| "learning_rate": 5.882352941176471e-05, |
| "loss": 0.0011, |
| "step": 480 |
| }, |
| { |
| "epoch": 2.88, |
| "learning_rate": 5.588235294117647e-05, |
| "loss": 0.0011, |
| "step": 490 |
| }, |
| { |
| "epoch": 2.94, |
| "learning_rate": 5.294117647058824e-05, |
| "loss": 0.0011, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.94, |
| "eval_accuracy": 1.0, |
| "eval_f1": 1.0, |
| "eval_loss": 0.0013941257493570447, |
| "eval_precision": 1.0, |
| "eval_recall": 1.0, |
| "eval_runtime": 12.2299, |
| "eval_samples_per_second": 47.752, |
| "eval_steps_per_second": 5.969, |
| "step": 500 |
| }, |
| { |
| "epoch": 3.0, |
| "learning_rate": 5e-05, |
| "loss": 0.001, |
| "step": 510 |
| }, |
| { |
| "epoch": 3.06, |
| "learning_rate": 4.705882352941177e-05, |
| "loss": 0.001, |
| "step": 520 |
| }, |
| { |
| "epoch": 3.12, |
| "learning_rate": 4.411764705882353e-05, |
| "loss": 0.0011, |
| "step": 530 |
| }, |
| { |
| "epoch": 3.18, |
| "learning_rate": 4.11764705882353e-05, |
| "loss": 0.001, |
| "step": 540 |
| }, |
| { |
| "epoch": 3.24, |
| "learning_rate": 3.8235294117647055e-05, |
| "loss": 0.001, |
| "step": 550 |
| }, |
| { |
| "epoch": 3.24, |
| "eval_accuracy": 1.0, |
| "eval_f1": 1.0, |
| "eval_loss": 0.001295336289331317, |
| "eval_precision": 1.0, |
| "eval_recall": 1.0, |
| "eval_runtime": 12.262, |
| "eval_samples_per_second": 47.627, |
| "eval_steps_per_second": 5.953, |
| "step": 550 |
| }, |
| { |
| "epoch": 3.29, |
| "learning_rate": 3.529411764705883e-05, |
| "loss": 0.001, |
| "step": 560 |
| }, |
| { |
| "epoch": 3.35, |
| "learning_rate": 3.235294117647059e-05, |
| "loss": 0.001, |
| "step": 570 |
| }, |
| { |
| "epoch": 3.41, |
| "learning_rate": 2.9411764705882354e-05, |
| "loss": 0.001, |
| "step": 580 |
| }, |
| { |
| "epoch": 3.47, |
| "learning_rate": 2.647058823529412e-05, |
| "loss": 0.0009, |
| "step": 590 |
| }, |
| { |
| "epoch": 3.53, |
| "learning_rate": 2.3529411764705884e-05, |
| "loss": 0.0009, |
| "step": 600 |
| }, |
| { |
| "epoch": 3.53, |
| "eval_accuracy": 1.0, |
| "eval_f1": 1.0, |
| "eval_loss": 0.0012326836585998535, |
| "eval_precision": 1.0, |
| "eval_recall": 1.0, |
| "eval_runtime": 12.2022, |
| "eval_samples_per_second": 47.86, |
| "eval_steps_per_second": 5.983, |
| "step": 600 |
| }, |
| { |
| "epoch": 3.59, |
| "learning_rate": 2.058823529411765e-05, |
| "loss": 0.0009, |
| "step": 610 |
| }, |
| { |
| "epoch": 3.65, |
| "learning_rate": 1.7647058823529414e-05, |
| "loss": 0.0009, |
| "step": 620 |
| }, |
| { |
| "epoch": 3.71, |
| "learning_rate": 1.4705882352941177e-05, |
| "loss": 0.0009, |
| "step": 630 |
| }, |
| { |
| "epoch": 3.76, |
| "learning_rate": 1.1764705882352942e-05, |
| "loss": 0.0009, |
| "step": 640 |
| }, |
| { |
| "epoch": 3.82, |
| "learning_rate": 8.823529411764707e-06, |
| "loss": 0.0009, |
| "step": 650 |
| }, |
| { |
| "epoch": 3.82, |
| "eval_accuracy": 1.0, |
| "eval_f1": 1.0, |
| "eval_loss": 0.0012013022787868977, |
| "eval_precision": 1.0, |
| "eval_recall": 1.0, |
| "eval_runtime": 12.3868, |
| "eval_samples_per_second": 47.147, |
| "eval_steps_per_second": 5.893, |
| "step": 650 |
| }, |
| { |
| "epoch": 3.88, |
| "learning_rate": 5.882352941176471e-06, |
| "loss": 0.0009, |
| "step": 660 |
| }, |
| { |
| "epoch": 3.94, |
| "learning_rate": 2.9411764705882355e-06, |
| "loss": 0.0009, |
| "step": 670 |
| }, |
| { |
| "epoch": 4.0, |
| "learning_rate": 0.0, |
| "loss": 0.001, |
| "step": 680 |
| }, |
| { |
| "epoch": 4.0, |
| "step": 680, |
| "total_flos": 6.18454266369137e+18, |
| "train_loss": 0.020127322464523947, |
| "train_runtime": 978.8347, |
| "train_samples_per_second": 11.091, |
| "train_steps_per_second": 0.695 |
| } |
| ], |
| "max_steps": 680, |
| "num_train_epochs": 4, |
| "total_flos": 6.18454266369137e+18, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
