{ "best_metric": 1.7676063776016235, "best_model_checkpoint": "detr-r50-cd45rb-2ah-6l-corrected\\checkpoint-115150", "epoch": 25.0, "global_step": 115150, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "learning_rate": 9.002171081198437e-06, "loss": 3.4312, "step": 4606 }, { "epoch": 1.0, "eval_loss": 2.442598342895508, "eval_runtime": 204.3572, "eval_samples_per_second": 8.715, "eval_steps_per_second": 1.091, "step": 4606 }, { "epoch": 2.0, "learning_rate": 8.002388189318281e-06, "loss": 2.9035, "step": 9212 }, { "epoch": 2.0, "eval_loss": 2.1671295166015625, "eval_runtime": 204.0022, "eval_samples_per_second": 8.73, "eval_steps_per_second": 1.093, "step": 9212 }, { "epoch": 3.0, "learning_rate": 7.0028224055579685e-06, "loss": 2.6788, "step": 13818 }, { "epoch": 3.0, "eval_loss": 2.0924932956695557, "eval_runtime": 203.9348, "eval_samples_per_second": 8.733, "eval_steps_per_second": 1.093, "step": 13818 }, { "epoch": 4.0, "learning_rate": 6.003256621797656e-06, "loss": 2.5847, "step": 18424 }, { "epoch": 4.0, "eval_loss": 2.021491765975952, "eval_runtime": 201.6799, "eval_samples_per_second": 8.831, "eval_steps_per_second": 1.106, "step": 18424 }, { "epoch": 5.0, "learning_rate": 5.0034737299175e-06, "loss": 2.5075, "step": 23030 }, { "epoch": 5.0, "eval_loss": 1.9688512086868286, "eval_runtime": 205.2053, "eval_samples_per_second": 8.679, "eval_steps_per_second": 1.087, "step": 23030 }, { "epoch": 6.0, "learning_rate": 4.0041250542770305e-06, "loss": 2.445, "step": 27636 }, { "epoch": 6.0, "eval_loss": 1.9527798891067505, "eval_runtime": 205.1637, "eval_samples_per_second": 8.681, "eval_steps_per_second": 1.087, "step": 27636 }, { "epoch": 7.0, "learning_rate": 3.0045592705167175e-06, "loss": 2.4078, "step": 32242 }, { "epoch": 7.0, "eval_loss": 1.9058469533920288, "eval_runtime": 205.1138, "eval_samples_per_second": 8.683, "eval_steps_per_second": 1.087, "step": 32242 }, { "epoch": 8.0, "learning_rate": 2.004993486756405e-06, "loss": 2.3794, "step": 36848 }, { "epoch": 8.0, "eval_loss": 1.888853907585144, "eval_runtime": 205.5976, "eval_samples_per_second": 8.663, "eval_steps_per_second": 1.085, "step": 36848 }, { "epoch": 9.0, "learning_rate": 1.0049934867564047e-06, "loss": 2.3572, "step": 41454 }, { "epoch": 9.0, "eval_loss": 1.875107765197754, "eval_runtime": 202.1582, "eval_samples_per_second": 8.81, "eval_steps_per_second": 1.103, "step": 41454 }, { "epoch": 10.0, "learning_rate": 5.427702996092055e-09, "loss": 2.3374, "step": 46060 }, { "epoch": 10.0, "eval_loss": 1.8622814416885376, "eval_runtime": 202.8571, "eval_samples_per_second": 8.78, "eval_steps_per_second": 1.099, "step": 46060 }, { "epoch": 11.0, "learning_rate": 2.6707193515704154e-06, "loss": 2.3423, "step": 50666 }, { "epoch": 11.0, "eval_loss": 1.8636469841003418, "eval_runtime": 238.0394, "eval_samples_per_second": 7.482, "eval_steps_per_second": 0.937, "step": 50666 }, { "epoch": 12.0, "learning_rate": 2.004052684903749e-06, "loss": 2.3292, "step": 55272 }, { "epoch": 12.0, "eval_loss": 1.8500159978866577, "eval_runtime": 233.3265, "eval_samples_per_second": 7.633, "eval_steps_per_second": 0.956, "step": 55272 }, { "epoch": 13.0, "learning_rate": 1.3378202344767697e-06, "loss": 2.3078, "step": 59878 }, { "epoch": 13.0, "eval_loss": 1.835439682006836, "eval_runtime": 237.6682, "eval_samples_per_second": 7.494, "eval_steps_per_second": 0.938, "step": 59878 }, { "epoch": 14.0, "learning_rate": 6.712983065566653e-07, "loss": 2.2958, "step": 64484 }, { "epoch": 14.0, "eval_loss": 1.8350275754928589, "eval_runtime": 235.9233, "eval_samples_per_second": 7.549, "eval_steps_per_second": 0.945, "step": 64484 }, { "epoch": 15.0, "learning_rate": 4.7763786365610075e-09, "loss": 2.2779, "step": 69090 }, { "epoch": 15.0, "eval_loss": 1.8243227005004883, "eval_runtime": 233.2527, "eval_samples_per_second": 7.635, "eval_steps_per_second": 0.956, "step": 69090 }, { "epoch": 16.0, "learning_rate": 3.603039513677812e-06, "loss": 2.3118, "step": 73696 }, { "epoch": 16.0, "eval_loss": 1.8304235935211182, "eval_runtime": 397.6353, "eval_samples_per_second": 4.479, "eval_steps_per_second": 0.561, "step": 73696 }, { "epoch": 17.0, "learning_rate": 3.2033000434216244e-06, "loss": 2.2927, "step": 78302 }, { "epoch": 17.0, "eval_loss": 1.8242826461791992, "eval_runtime": 232.9531, "eval_samples_per_second": 7.645, "eval_steps_per_second": 0.957, "step": 78302 }, { "epoch": 18.0, "learning_rate": 2.803386886669562e-06, "loss": 2.2726, "step": 82908 }, { "epoch": 18.0, "eval_loss": 1.816089153289795, "eval_runtime": 231.9979, "eval_samples_per_second": 7.677, "eval_steps_per_second": 0.961, "step": 82908 }, { "epoch": 19.0, "learning_rate": 2.403473729917499e-06, "loss": 2.2546, "step": 87514 }, { "epoch": 19.0, "eval_loss": 1.810555100440979, "eval_runtime": 234.519, "eval_samples_per_second": 7.594, "eval_steps_per_second": 0.951, "step": 87514 }, { "epoch": 20.0, "learning_rate": 2.0035605731654363e-06, "loss": 2.2394, "step": 92120 }, { "epoch": 20.0, "eval_loss": 1.794918417930603, "eval_runtime": 235.0533, "eval_samples_per_second": 7.577, "eval_steps_per_second": 0.949, "step": 92120 }, { "epoch": 21.0, "learning_rate": 1.603821102909249e-06, "loss": 2.237, "step": 96726 }, { "epoch": 21.0, "eval_loss": 1.7747350931167603, "eval_runtime": 690.7728, "eval_samples_per_second": 2.578, "eval_steps_per_second": 0.323, "step": 96726 }, { "epoch": 22.0, "learning_rate": 1.203821102909249e-06, "loss": 2.2158, "step": 101332 }, { "epoch": 22.0, "eval_loss": 1.7796499729156494, "eval_runtime": 699.9768, "eval_samples_per_second": 2.544, "eval_steps_per_second": 0.319, "step": 101332 }, { "epoch": 23.0, "learning_rate": 8.039947894051238e-07, "loss": 2.2069, "step": 105938 }, { "epoch": 23.0, "eval_loss": 1.7756770849227905, "eval_runtime": 1199.5055, "eval_samples_per_second": 1.485, "eval_steps_per_second": 0.186, "step": 105938 }, { "epoch": 24.0, "learning_rate": 4.0425531914893625e-07, "loss": 2.2061, "step": 110544 }, { "epoch": 24.0, "eval_loss": 1.7745826244354248, "eval_runtime": 1335.191, "eval_samples_per_second": 1.334, "eval_steps_per_second": 0.167, "step": 110544 }, { "epoch": 25.0, "learning_rate": 4.342162396873643e-09, "loss": 2.1974, "step": 115150 }, { "epoch": 25.0, "eval_loss": 1.7676063776016235, "eval_runtime": 1343.9076, "eval_samples_per_second": 1.325, "eval_steps_per_second": 0.166, "step": 115150 }, { "epoch": 25.0, "step": 115150, "total_flos": 2.2008141195206407e+20, "train_loss": 0.8973775459319366, "train_runtime": 77323.2871, "train_samples_per_second": 5.956, "train_steps_per_second": 1.489 } ], "max_steps": 115150, "num_train_epochs": 25, "total_flos": 2.2008141195206407e+20, "trial_name": null, "trial_params": null }