| { | |
| "best_metric": 0.2543001174926758, | |
| "best_model_checkpoint": "/workspace/training_output/perspectives-vit-large-patch16-224-in21k/checkpoint-12168", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 15210, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.9342537804076265e-05, | |
| "loss": 0.9137, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1.8685075608152533e-05, | |
| "loss": 0.5533, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.80276134122288e-05, | |
| "loss": 0.5205, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.7370151216305063e-05, | |
| "loss": 0.4745, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.671268902038133e-05, | |
| "loss": 0.4605, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.6055226824457594e-05, | |
| "loss": 0.4164, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9024219841639497, | |
| "eval_loss": 0.28684180974960327, | |
| "eval_runtime": 335.4974, | |
| "eval_samples_per_second": 12.799, | |
| "eval_steps_per_second": 1.601, | |
| "step": 3042 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.539776462853386e-05, | |
| "loss": 0.3762, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.4740302432610125e-05, | |
| "loss": 0.3825, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 1.4082840236686392e-05, | |
| "loss": 0.3543, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 1.3425378040762658e-05, | |
| "loss": 0.3748, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 1.2767915844838923e-05, | |
| "loss": 0.3462, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 1.2110453648915189e-05, | |
| "loss": 0.3391, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.904052165812762, | |
| "eval_loss": 0.30549806356430054, | |
| "eval_runtime": 334.7146, | |
| "eval_samples_per_second": 12.829, | |
| "eval_steps_per_second": 1.604, | |
| "step": 6084 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 1.1452991452991454e-05, | |
| "loss": 0.3078, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.0795529257067721e-05, | |
| "loss": 0.2976, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1.0138067061143987e-05, | |
| "loss": 0.3078, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 9.48060486522025e-06, | |
| "loss": 0.2773, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 8.823142669296516e-06, | |
| "loss": 0.2851, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 8.165680473372781e-06, | |
| "loss": 0.2836, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9180251513740102, | |
| "eval_loss": 0.30707448720932007, | |
| "eval_runtime": 337.2814, | |
| "eval_samples_per_second": 12.731, | |
| "eval_steps_per_second": 1.592, | |
| "step": 9126 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 7.508218277449047e-06, | |
| "loss": 0.2507, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 6.850756081525313e-06, | |
| "loss": 0.2584, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 6.193293885601579e-06, | |
| "loss": 0.27, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 5.535831689677844e-06, | |
| "loss": 0.2792, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 4.878369493754109e-06, | |
| "loss": 0.2335, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 4.220907297830375e-06, | |
| "loss": 0.2292, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9315323707498836, | |
| "eval_loss": 0.2543001174926758, | |
| "eval_runtime": 337.9988, | |
| "eval_samples_per_second": 12.704, | |
| "eval_steps_per_second": 1.589, | |
| "step": 12168 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 3.563445101906641e-06, | |
| "loss": 0.2234, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 2.9059829059829063e-06, | |
| "loss": 0.2069, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 2.2485207100591717e-06, | |
| "loss": 0.2118, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 1.5910585141354374e-06, | |
| "loss": 0.2041, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 9.335963182117029e-07, | |
| "loss": 0.2159, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 2.7613412228796843e-07, | |
| "loss": 0.1842, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.936190032603633, | |
| "eval_loss": 0.256198912858963, | |
| "eval_runtime": 350.5885, | |
| "eval_samples_per_second": 12.248, | |
| "eval_steps_per_second": 1.532, | |
| "step": 15210 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 15210, | |
| "total_flos": 3.332544438628823e+19, | |
| "train_loss": 0.3327771868382842, | |
| "train_runtime": 12171.5283, | |
| "train_samples_per_second": 9.995, | |
| "train_steps_per_second": 1.25 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 15210, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 3.332544438628823e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |