| { | |
| "best_metric": 0.18186990916728973, | |
| "best_model_checkpoint": "face_obstruction_image_detection/checkpoint-11556", | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 11556, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 9.648162627052385e-06, | |
| "loss": 0.8123, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 9.257232212666147e-06, | |
| "loss": 0.3519, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8846678355737385, | |
| "eval_loss": 0.2906096577644348, | |
| "eval_runtime": 140.8808, | |
| "eval_samples_per_second": 72.87, | |
| "eval_steps_per_second": 9.114, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 8.866301798279907e-06, | |
| "loss": 0.2891, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 8.475371383893667e-06, | |
| "loss": 0.2514, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 8.08444096950743e-06, | |
| "loss": 0.2335, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8915838690824079, | |
| "eval_loss": 0.2299681454896927, | |
| "eval_runtime": 140.059, | |
| "eval_samples_per_second": 73.298, | |
| "eval_steps_per_second": 9.168, | |
| "step": 2568 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 7.693510555121188e-06, | |
| "loss": 0.2103, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 7.30258014073495e-06, | |
| "loss": 0.2044, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.891291642314436, | |
| "eval_loss": 0.2096417248249054, | |
| "eval_runtime": 141.1058, | |
| "eval_samples_per_second": 72.754, | |
| "eval_steps_per_second": 9.1, | |
| "step": 3852 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 6.91164972634871e-06, | |
| "loss": 0.1924, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 6.520719311962471e-06, | |
| "loss": 0.1837, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 6.129788897576232e-06, | |
| "loss": 0.1807, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8915838690824079, | |
| "eval_loss": 0.19587308168411255, | |
| "eval_runtime": 137.5403, | |
| "eval_samples_per_second": 74.64, | |
| "eval_steps_per_second": 9.335, | |
| "step": 5136 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 5.7388584831899925e-06, | |
| "loss": 0.1679, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 5.347928068803754e-06, | |
| "loss": 0.1686, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8925579583089811, | |
| "eval_loss": 0.18770596385002136, | |
| "eval_runtime": 139.4839, | |
| "eval_samples_per_second": 73.6, | |
| "eval_steps_per_second": 9.205, | |
| "step": 6420 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 4.956997654417514e-06, | |
| "loss": 0.1696, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 4.566067240031275e-06, | |
| "loss": 0.165, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 4.175136825645035e-06, | |
| "loss": 0.1566, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8901227352425483, | |
| "eval_loss": 0.1865508109331131, | |
| "eval_runtime": 140.3224, | |
| "eval_samples_per_second": 73.16, | |
| "eval_steps_per_second": 9.15, | |
| "step": 7704 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 3.7842064112587963e-06, | |
| "loss": 0.1586, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 3.393275996872557e-06, | |
| "loss": 0.1556, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8917786869277226, | |
| "eval_loss": 0.18248248100280762, | |
| "eval_runtime": 141.8634, | |
| "eval_samples_per_second": 72.365, | |
| "eval_steps_per_second": 9.051, | |
| "step": 8988 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 3.002345582486318e-06, | |
| "loss": 0.1591, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "learning_rate": 2.6114151681000783e-06, | |
| "loss": 0.1544, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 2.220484753713839e-06, | |
| "loss": 0.1491, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.893142411844925, | |
| "eval_loss": 0.18403713405132294, | |
| "eval_runtime": 143.6244, | |
| "eval_samples_per_second": 71.478, | |
| "eval_steps_per_second": 8.94, | |
| "step": 10272 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 1.8295543393276e-06, | |
| "loss": 0.1522, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 1.4386239249413605e-06, | |
| "loss": 0.1468, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 1.0476935105551214e-06, | |
| "loss": 0.1477, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8919735047730372, | |
| "eval_loss": 0.18186990916728973, | |
| "eval_runtime": 143.3138, | |
| "eval_samples_per_second": 71.633, | |
| "eval_steps_per_second": 8.959, | |
| "step": 11556 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 12840, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 2.864020628943323e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |