| { | |
| "best_metric": 3.882021427154541, | |
| "best_model_checkpoint": "detr-r50-mist1-bg-2ah-6l/checkpoint-2760", | |
| "epoch": 25.0, | |
| "eval_steps": 500, | |
| "global_step": 2875, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.624347826086956e-06, | |
| "loss": 4.6721, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 5.003213882446289, | |
| "eval_runtime": 5.6326, | |
| "eval_samples_per_second": 7.101, | |
| "eval_steps_per_second": 0.888, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.224347826086957e-06, | |
| "loss": 4.4438, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 4.679709434509277, | |
| "eval_runtime": 5.7025, | |
| "eval_samples_per_second": 7.015, | |
| "eval_steps_per_second": 0.877, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.824347826086957e-06, | |
| "loss": 4.2953, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 4.702702522277832, | |
| "eval_runtime": 5.7246, | |
| "eval_samples_per_second": 6.987, | |
| "eval_steps_per_second": 0.873, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8.424347826086958e-06, | |
| "loss": 4.3899, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 5.431577682495117, | |
| "eval_runtime": 5.6954, | |
| "eval_samples_per_second": 7.023, | |
| "eval_steps_per_second": 0.878, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 8.024347826086956e-06, | |
| "loss": 4.3184, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 4.41250467300415, | |
| "eval_runtime": 5.6729, | |
| "eval_samples_per_second": 7.051, | |
| "eval_steps_per_second": 0.881, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7.624347826086958e-06, | |
| "loss": 4.2749, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 4.161060810089111, | |
| "eval_runtime": 5.6652, | |
| "eval_samples_per_second": 7.061, | |
| "eval_steps_per_second": 0.883, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 7.227826086956522e-06, | |
| "loss": 4.2153, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 4.672310829162598, | |
| "eval_runtime": 5.6197, | |
| "eval_samples_per_second": 7.118, | |
| "eval_steps_per_second": 0.89, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 6.827826086956523e-06, | |
| "loss": 4.0788, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 4.12662410736084, | |
| "eval_runtime": 5.6637, | |
| "eval_samples_per_second": 7.062, | |
| "eval_steps_per_second": 0.883, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 6.427826086956523e-06, | |
| "loss": 4.0752, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 4.052855014801025, | |
| "eval_runtime": 5.641, | |
| "eval_samples_per_second": 7.091, | |
| "eval_steps_per_second": 0.886, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 6.027826086956522e-06, | |
| "loss": 4.0073, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 4.448337078094482, | |
| "eval_runtime": 5.6934, | |
| "eval_samples_per_second": 7.026, | |
| "eval_steps_per_second": 0.878, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 5.627826086956523e-06, | |
| "loss": 4.011, | |
| "step": 1265 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 4.200244426727295, | |
| "eval_runtime": 5.6368, | |
| "eval_samples_per_second": 7.096, | |
| "eval_steps_per_second": 0.887, | |
| "step": 1265 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 5.227826086956522e-06, | |
| "loss": 3.9993, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 4.2449798583984375, | |
| "eval_runtime": 5.6784, | |
| "eval_samples_per_second": 7.044, | |
| "eval_steps_per_second": 0.881, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 4.827826086956522e-06, | |
| "loss": 4.0028, | |
| "step": 1495 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 4.170295715332031, | |
| "eval_runtime": 5.8028, | |
| "eval_samples_per_second": 6.893, | |
| "eval_steps_per_second": 0.862, | |
| "step": 1495 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 4.427826086956522e-06, | |
| "loss": 3.9572, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 4.186112880706787, | |
| "eval_runtime": 5.7389, | |
| "eval_samples_per_second": 6.97, | |
| "eval_steps_per_second": 0.871, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 4.027826086956522e-06, | |
| "loss": 3.9009, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 4.02849006652832, | |
| "eval_runtime": 5.8134, | |
| "eval_samples_per_second": 6.881, | |
| "eval_steps_per_second": 0.86, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 3.627826086956522e-06, | |
| "loss": 3.9173, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 4.06730842590332, | |
| "eval_runtime": 5.7961, | |
| "eval_samples_per_second": 6.901, | |
| "eval_steps_per_second": 0.863, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 3.227826086956522e-06, | |
| "loss": 3.8884, | |
| "step": 1955 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 3.9875235557556152, | |
| "eval_runtime": 5.7593, | |
| "eval_samples_per_second": 6.945, | |
| "eval_steps_per_second": 0.868, | |
| "step": 1955 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 2.8278260869565224e-06, | |
| "loss": 3.8415, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 4.106224536895752, | |
| "eval_runtime": 5.7793, | |
| "eval_samples_per_second": 6.921, | |
| "eval_steps_per_second": 0.865, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 2.427826086956522e-06, | |
| "loss": 3.8132, | |
| "step": 2185 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 4.049378871917725, | |
| "eval_runtime": 5.6624, | |
| "eval_samples_per_second": 7.064, | |
| "eval_steps_per_second": 0.883, | |
| "step": 2185 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 2.027826086956522e-06, | |
| "loss": 3.8297, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 4.011868476867676, | |
| "eval_runtime": 5.745, | |
| "eval_samples_per_second": 6.963, | |
| "eval_steps_per_second": 0.87, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 1.6278260869565218e-06, | |
| "loss": 3.8262, | |
| "step": 2415 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 3.953753709793091, | |
| "eval_runtime": 5.7295, | |
| "eval_samples_per_second": 6.981, | |
| "eval_steps_per_second": 0.873, | |
| "step": 2415 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 1.2278260869565218e-06, | |
| "loss": 3.8045, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 3.950028657913208, | |
| "eval_runtime": 5.8404, | |
| "eval_samples_per_second": 6.849, | |
| "eval_steps_per_second": 0.856, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 8.278260869565217e-07, | |
| "loss": 3.8067, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 3.92639422416687, | |
| "eval_runtime": 5.7491, | |
| "eval_samples_per_second": 6.958, | |
| "eval_steps_per_second": 0.87, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 4.278260869565218e-07, | |
| "loss": 3.7651, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 3.882021427154541, | |
| "eval_runtime": 5.8762, | |
| "eval_samples_per_second": 6.807, | |
| "eval_steps_per_second": 0.851, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 2.7826086956521744e-08, | |
| "loss": 3.756, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 3.9051177501678467, | |
| "eval_runtime": 5.7364, | |
| "eval_samples_per_second": 6.973, | |
| "eval_steps_per_second": 0.872, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "step": 2875, | |
| "total_flos": 5.49474551424e+18, | |
| "train_loss": 4.035632897418478, | |
| "train_runtime": 2307.326, | |
| "train_samples_per_second": 4.984, | |
| "train_steps_per_second": 1.246 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2875, | |
| "num_train_epochs": 25, | |
| "save_steps": 500, | |
| "total_flos": 5.49474551424e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |