| { |
| "best_metric": 0.9085714285714286, |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DAV61/checkpoint-210", |
| "epoch": 40.51282051282051, |
| "eval_steps": 500, |
| "global_step": 405, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 4.095645427703857, |
| "learning_rate": 9.756097560975611e-06, |
| "loss": 1.1254, |
| "step": 10 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.48, |
| "eval_loss": 1.0551393032073975, |
| "eval_runtime": 2.0191, |
| "eval_samples_per_second": 86.673, |
| "eval_steps_per_second": 5.448, |
| "step": 10 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 3.2705843448638916, |
| "learning_rate": 1.9512195121951222e-05, |
| "loss": 0.967, |
| "step": 20 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.45714285714285713, |
| "eval_loss": 0.976972758769989, |
| "eval_runtime": 1.8626, |
| "eval_samples_per_second": 93.954, |
| "eval_steps_per_second": 5.906, |
| "step": 20 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 4.272066116333008, |
| "learning_rate": 2.926829268292683e-05, |
| "loss": 0.8537, |
| "step": 30 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.7085714285714285, |
| "eval_loss": 0.7128326296806335, |
| "eval_runtime": 1.8579, |
| "eval_samples_per_second": 94.192, |
| "eval_steps_per_second": 5.921, |
| "step": 30 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 5.0191121101379395, |
| "learning_rate": 3.9024390243902444e-05, |
| "loss": 0.6489, |
| "step": 40 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.7371428571428571, |
| "eval_loss": 0.5167368650436401, |
| "eval_runtime": 1.8933, |
| "eval_samples_per_second": 92.43, |
| "eval_steps_per_second": 5.81, |
| "step": 40 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 8.822305679321289, |
| "learning_rate": 3.901098901098901e-05, |
| "loss": 0.567, |
| "step": 50 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8, |
| "eval_loss": 0.3952338397502899, |
| "eval_runtime": 2.4771, |
| "eval_samples_per_second": 70.647, |
| "eval_steps_per_second": 4.441, |
| "step": 50 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 9.881946563720703, |
| "learning_rate": 3.802197802197802e-05, |
| "loss": 0.4724, |
| "step": 60 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8571428571428571, |
| "eval_loss": 0.37049734592437744, |
| "eval_runtime": 2.5683, |
| "eval_samples_per_second": 68.139, |
| "eval_steps_per_second": 4.283, |
| "step": 60 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 6.562332630157471, |
| "learning_rate": 3.692307692307693e-05, |
| "loss": 0.4106, |
| "step": 70 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8742857142857143, |
| "eval_loss": 0.3075900971889496, |
| "eval_runtime": 1.9438, |
| "eval_samples_per_second": 90.03, |
| "eval_steps_per_second": 5.659, |
| "step": 70 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 7.867884635925293, |
| "learning_rate": 3.582417582417583e-05, |
| "loss": 0.3853, |
| "step": 80 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8914285714285715, |
| "eval_loss": 0.31545019149780273, |
| "eval_runtime": 1.8617, |
| "eval_samples_per_second": 94.0, |
| "eval_steps_per_second": 5.909, |
| "step": 80 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 7.147304534912109, |
| "learning_rate": 3.4725274725274726e-05, |
| "loss": 0.3369, |
| "step": 90 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.8514285714285714, |
| "eval_loss": 0.2907550632953644, |
| "eval_runtime": 1.8589, |
| "eval_samples_per_second": 94.14, |
| "eval_steps_per_second": 5.917, |
| "step": 90 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 8.581195831298828, |
| "learning_rate": 3.362637362637363e-05, |
| "loss": 0.3345, |
| "step": 100 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.8457142857142858, |
| "eval_loss": 0.30341535806655884, |
| "eval_runtime": 1.9147, |
| "eval_samples_per_second": 91.399, |
| "eval_steps_per_second": 5.745, |
| "step": 100 |
| }, |
| { |
| "epoch": 11.0, |
| "grad_norm": 4.6522135734558105, |
| "learning_rate": 3.252747252747253e-05, |
| "loss": 0.3092, |
| "step": 110 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_accuracy": 0.88, |
| "eval_loss": 0.25752559304237366, |
| "eval_runtime": 1.8952, |
| "eval_samples_per_second": 92.338, |
| "eval_steps_per_second": 5.804, |
| "step": 110 |
| }, |
| { |
| "epoch": 12.0, |
| "grad_norm": 7.967129230499268, |
| "learning_rate": 3.142857142857143e-05, |
| "loss": 0.3027, |
| "step": 120 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_accuracy": 0.8971428571428571, |
| "eval_loss": 0.248607337474823, |
| "eval_runtime": 2.7567, |
| "eval_samples_per_second": 63.483, |
| "eval_steps_per_second": 3.99, |
| "step": 120 |
| }, |
| { |
| "epoch": 13.0, |
| "grad_norm": 4.966682434082031, |
| "learning_rate": 3.0329670329670332e-05, |
| "loss": 0.2945, |
| "step": 130 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_accuracy": 0.8628571428571429, |
| "eval_loss": 0.2955775558948517, |
| "eval_runtime": 1.9442, |
| "eval_samples_per_second": 90.013, |
| "eval_steps_per_second": 5.658, |
| "step": 130 |
| }, |
| { |
| "epoch": 14.0, |
| "grad_norm": 4.427709102630615, |
| "learning_rate": 2.923076923076923e-05, |
| "loss": 0.254, |
| "step": 140 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_accuracy": 0.8971428571428571, |
| "eval_loss": 0.2787053883075714, |
| "eval_runtime": 1.8912, |
| "eval_samples_per_second": 92.533, |
| "eval_steps_per_second": 5.816, |
| "step": 140 |
| }, |
| { |
| "epoch": 15.0, |
| "grad_norm": 4.446314334869385, |
| "learning_rate": 2.8131868131868136e-05, |
| "loss": 0.2476, |
| "step": 150 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_accuracy": 0.9028571428571428, |
| "eval_loss": 0.2594496011734009, |
| "eval_runtime": 1.9258, |
| "eval_samples_per_second": 90.874, |
| "eval_steps_per_second": 5.712, |
| "step": 150 |
| }, |
| { |
| "epoch": 16.0, |
| "grad_norm": 6.341131687164307, |
| "learning_rate": 2.7032967032967037e-05, |
| "loss": 0.2058, |
| "step": 160 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_accuracy": 0.8971428571428571, |
| "eval_loss": 0.3091249167919159, |
| "eval_runtime": 1.9651, |
| "eval_samples_per_second": 89.054, |
| "eval_steps_per_second": 5.598, |
| "step": 160 |
| }, |
| { |
| "epoch": 17.0, |
| "grad_norm": 4.9936041831970215, |
| "learning_rate": 2.593406593406594e-05, |
| "loss": 0.2254, |
| "step": 170 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_accuracy": 0.8971428571428571, |
| "eval_loss": 0.27779555320739746, |
| "eval_runtime": 1.9764, |
| "eval_samples_per_second": 88.543, |
| "eval_steps_per_second": 5.566, |
| "step": 170 |
| }, |
| { |
| "epoch": 18.0, |
| "grad_norm": 8.642626762390137, |
| "learning_rate": 2.4835164835164837e-05, |
| "loss": 0.2165, |
| "step": 180 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_accuracy": 0.88, |
| "eval_loss": 0.3093833029270172, |
| "eval_runtime": 2.7814, |
| "eval_samples_per_second": 62.918, |
| "eval_steps_per_second": 3.955, |
| "step": 180 |
| }, |
| { |
| "epoch": 19.0, |
| "grad_norm": 4.932384490966797, |
| "learning_rate": 2.373626373626374e-05, |
| "loss": 0.2501, |
| "step": 190 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_accuracy": 0.8914285714285715, |
| "eval_loss": 0.25359097123146057, |
| "eval_runtime": 1.9572, |
| "eval_samples_per_second": 89.415, |
| "eval_steps_per_second": 5.62, |
| "step": 190 |
| }, |
| { |
| "epoch": 20.0, |
| "grad_norm": 3.465360641479492, |
| "learning_rate": 2.263736263736264e-05, |
| "loss": 0.2016, |
| "step": 200 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_accuracy": 0.9028571428571428, |
| "eval_loss": 0.2651709318161011, |
| "eval_runtime": 1.8914, |
| "eval_samples_per_second": 92.525, |
| "eval_steps_per_second": 5.816, |
| "step": 200 |
| }, |
| { |
| "epoch": 21.0, |
| "grad_norm": 6.498570919036865, |
| "learning_rate": 2.153846153846154e-05, |
| "loss": 0.1791, |
| "step": 210 |
| }, |
| { |
| "epoch": 21.0, |
| "eval_accuracy": 0.9085714285714286, |
| "eval_loss": 0.24487070739269257, |
| "eval_runtime": 1.9592, |
| "eval_samples_per_second": 89.322, |
| "eval_steps_per_second": 5.615, |
| "step": 210 |
| }, |
| { |
| "epoch": 22.0, |
| "grad_norm": 7.660604000091553, |
| "learning_rate": 2.043956043956044e-05, |
| "loss": 0.1401, |
| "step": 220 |
| }, |
| { |
| "epoch": 22.0, |
| "eval_accuracy": 0.8742857142857143, |
| "eval_loss": 0.3205382823944092, |
| "eval_runtime": 2.3493, |
| "eval_samples_per_second": 74.492, |
| "eval_steps_per_second": 4.682, |
| "step": 220 |
| }, |
| { |
| "epoch": 23.0, |
| "grad_norm": 3.7020044326782227, |
| "learning_rate": 1.9340659340659342e-05, |
| "loss": 0.17, |
| "step": 230 |
| }, |
| { |
| "epoch": 23.0, |
| "eval_accuracy": 0.8685714285714285, |
| "eval_loss": 0.34455186128616333, |
| "eval_runtime": 1.9524, |
| "eval_samples_per_second": 89.634, |
| "eval_steps_per_second": 5.634, |
| "step": 230 |
| }, |
| { |
| "epoch": 24.0, |
| "grad_norm": 4.02321195602417, |
| "learning_rate": 1.8241758241758244e-05, |
| "loss": 0.1665, |
| "step": 240 |
| }, |
| { |
| "epoch": 24.0, |
| "eval_accuracy": 0.8742857142857143, |
| "eval_loss": 0.30577367544174194, |
| "eval_runtime": 1.8968, |
| "eval_samples_per_second": 92.26, |
| "eval_steps_per_second": 5.799, |
| "step": 240 |
| }, |
| { |
| "epoch": 25.0, |
| "grad_norm": 3.3379478454589844, |
| "learning_rate": 1.7142857142857142e-05, |
| "loss": 0.1192, |
| "step": 250 |
| }, |
| { |
| "epoch": 25.0, |
| "eval_accuracy": 0.8742857142857143, |
| "eval_loss": 0.3537001609802246, |
| "eval_runtime": 1.9127, |
| "eval_samples_per_second": 91.495, |
| "eval_steps_per_second": 5.751, |
| "step": 250 |
| }, |
| { |
| "epoch": 26.0, |
| "grad_norm": 10.440197944641113, |
| "learning_rate": 1.6043956043956047e-05, |
| "loss": 0.1249, |
| "step": 260 |
| }, |
| { |
| "epoch": 26.0, |
| "eval_accuracy": 0.8742857142857143, |
| "eval_loss": 0.2928812801837921, |
| "eval_runtime": 2.692, |
| "eval_samples_per_second": 65.008, |
| "eval_steps_per_second": 4.086, |
| "step": 260 |
| }, |
| { |
| "epoch": 27.0, |
| "grad_norm": 7.9000935554504395, |
| "learning_rate": 1.4945054945054947e-05, |
| "loss": 0.143, |
| "step": 270 |
| }, |
| { |
| "epoch": 27.0, |
| "eval_accuracy": 0.8742857142857143, |
| "eval_loss": 0.3049885630607605, |
| "eval_runtime": 1.8758, |
| "eval_samples_per_second": 93.294, |
| "eval_steps_per_second": 5.864, |
| "step": 270 |
| }, |
| { |
| "epoch": 28.0, |
| "grad_norm": 4.663361549377441, |
| "learning_rate": 1.3846153846153847e-05, |
| "loss": 0.1809, |
| "step": 280 |
| }, |
| { |
| "epoch": 28.0, |
| "eval_accuracy": 0.8914285714285715, |
| "eval_loss": 0.2963820993900299, |
| "eval_runtime": 1.9174, |
| "eval_samples_per_second": 91.267, |
| "eval_steps_per_second": 5.737, |
| "step": 280 |
| }, |
| { |
| "epoch": 29.0, |
| "grad_norm": 4.101968765258789, |
| "learning_rate": 1.2747252747252747e-05, |
| "loss": 0.135, |
| "step": 290 |
| }, |
| { |
| "epoch": 29.0, |
| "eval_accuracy": 0.8742857142857143, |
| "eval_loss": 0.30466756224632263, |
| "eval_runtime": 1.9228, |
| "eval_samples_per_second": 91.012, |
| "eval_steps_per_second": 5.721, |
| "step": 290 |
| }, |
| { |
| "epoch": 30.0, |
| "grad_norm": 7.243139266967773, |
| "learning_rate": 1.164835164835165e-05, |
| "loss": 0.1235, |
| "step": 300 |
| }, |
| { |
| "epoch": 30.0, |
| "eval_accuracy": 0.8971428571428571, |
| "eval_loss": 0.3011048436164856, |
| "eval_runtime": 1.9213, |
| "eval_samples_per_second": 91.082, |
| "eval_steps_per_second": 5.725, |
| "step": 300 |
| }, |
| { |
| "epoch": 31.0, |
| "grad_norm": 8.738358497619629, |
| "learning_rate": 1.054945054945055e-05, |
| "loss": 0.1351, |
| "step": 310 |
| }, |
| { |
| "epoch": 31.0, |
| "eval_accuracy": 0.8914285714285715, |
| "eval_loss": 0.31473463773727417, |
| "eval_runtime": 2.1888, |
| "eval_samples_per_second": 79.951, |
| "eval_steps_per_second": 5.026, |
| "step": 310 |
| }, |
| { |
| "epoch": 32.0, |
| "grad_norm": 7.214813709259033, |
| "learning_rate": 9.450549450549452e-06, |
| "loss": 0.1124, |
| "step": 320 |
| }, |
| { |
| "epoch": 32.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.3122876286506653, |
| "eval_runtime": 1.9531, |
| "eval_samples_per_second": 89.602, |
| "eval_steps_per_second": 5.632, |
| "step": 320 |
| }, |
| { |
| "epoch": 33.0, |
| "grad_norm": 5.647735595703125, |
| "learning_rate": 8.351648351648353e-06, |
| "loss": 0.1122, |
| "step": 330 |
| }, |
| { |
| "epoch": 33.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.328876793384552, |
| "eval_runtime": 1.9152, |
| "eval_samples_per_second": 91.375, |
| "eval_steps_per_second": 5.744, |
| "step": 330 |
| }, |
| { |
| "epoch": 34.0, |
| "grad_norm": 7.423587322235107, |
| "learning_rate": 7.252747252747253e-06, |
| "loss": 0.1163, |
| "step": 340 |
| }, |
| { |
| "epoch": 34.0, |
| "eval_accuracy": 0.88, |
| "eval_loss": 0.3256431221961975, |
| "eval_runtime": 1.9355, |
| "eval_samples_per_second": 90.415, |
| "eval_steps_per_second": 5.683, |
| "step": 340 |
| }, |
| { |
| "epoch": 35.0, |
| "grad_norm": 2.3473429679870605, |
| "learning_rate": 6.153846153846155e-06, |
| "loss": 0.0885, |
| "step": 350 |
| }, |
| { |
| "epoch": 35.0, |
| "eval_accuracy": 0.8971428571428571, |
| "eval_loss": 0.3125978410243988, |
| "eval_runtime": 1.973, |
| "eval_samples_per_second": 88.695, |
| "eval_steps_per_second": 5.575, |
| "step": 350 |
| }, |
| { |
| "epoch": 36.0, |
| "grad_norm": 6.636789321899414, |
| "learning_rate": 5.054945054945055e-06, |
| "loss": 0.097, |
| "step": 360 |
| }, |
| { |
| "epoch": 36.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.342531681060791, |
| "eval_runtime": 2.7179, |
| "eval_samples_per_second": 64.389, |
| "eval_steps_per_second": 4.047, |
| "step": 360 |
| }, |
| { |
| "epoch": 37.0, |
| "grad_norm": 1.5700148344039917, |
| "learning_rate": 3.9560439560439565e-06, |
| "loss": 0.1036, |
| "step": 370 |
| }, |
| { |
| "epoch": 37.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.3220842778682709, |
| "eval_runtime": 1.9408, |
| "eval_samples_per_second": 90.168, |
| "eval_steps_per_second": 5.668, |
| "step": 370 |
| }, |
| { |
| "epoch": 38.0, |
| "grad_norm": 11.366103172302246, |
| "learning_rate": 2.8571428571428573e-06, |
| "loss": 0.1107, |
| "step": 380 |
| }, |
| { |
| "epoch": 38.0, |
| "eval_accuracy": 0.8971428571428571, |
| "eval_loss": 0.3597542345523834, |
| "eval_runtime": 1.9255, |
| "eval_samples_per_second": 90.887, |
| "eval_steps_per_second": 5.713, |
| "step": 380 |
| }, |
| { |
| "epoch": 39.0, |
| "grad_norm": 5.265808582305908, |
| "learning_rate": 1.7582417582417585e-06, |
| "loss": 0.0928, |
| "step": 390 |
| }, |
| { |
| "epoch": 39.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.33551570773124695, |
| "eval_runtime": 1.9453, |
| "eval_samples_per_second": 89.962, |
| "eval_steps_per_second": 5.655, |
| "step": 390 |
| }, |
| { |
| "epoch": 40.0, |
| "grad_norm": 8.759902000427246, |
| "learning_rate": 6.593406593406594e-07, |
| "loss": 0.1104, |
| "step": 400 |
| }, |
| { |
| "epoch": 40.0, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.3425438404083252, |
| "eval_runtime": 2.1768, |
| "eval_samples_per_second": 80.394, |
| "eval_steps_per_second": 5.053, |
| "step": 400 |
| }, |
| { |
| "epoch": 40.51282051282051, |
| "eval_accuracy": 0.8857142857142857, |
| "eval_loss": 0.3428290784358978, |
| "eval_runtime": 2.2062, |
| "eval_samples_per_second": 79.32, |
| "eval_steps_per_second": 4.986, |
| "step": 405 |
| }, |
| { |
| "epoch": 40.51282051282051, |
| "step": 405, |
| "total_flos": 8.172851197221274e+17, |
| "train_loss": 0.2773810105559267, |
| "train_runtime": 635.96, |
| "train_samples_per_second": 43.871, |
| "train_steps_per_second": 0.637 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 405, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 45, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 8.172851197221274e+17, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
