swin-tiny-patch4-window7-224-finetuned-teeth_dataset-finetuned-teeth_dataset-V2
/
trainer_state.json
| { | |
| "best_metric": 0.9260869565217391, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-teeth_dataset-finetuned-teeth_dataset-V2/checkpoint-135", | |
| "epoch": 40.0, | |
| "eval_steps": 500, | |
| "global_step": 150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.8391304347826087, | |
| "eval_loss": 1.1266765594482422, | |
| "eval_runtime": 299.2277, | |
| "eval_samples_per_second": 1.537, | |
| "eval_steps_per_second": 0.05, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_accuracy": 0.8304347826086956, | |
| "eval_loss": 1.0719478130340576, | |
| "eval_runtime": 4.3394, | |
| "eval_samples_per_second": 106.006, | |
| "eval_steps_per_second": 3.457, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 11.345915794372559, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.5709, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_accuracy": 0.8478260869565217, | |
| "eval_loss": 0.944671630859375, | |
| "eval_runtime": 4.3766, | |
| "eval_samples_per_second": 105.105, | |
| "eval_steps_per_second": 3.427, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8652173913043478, | |
| "eval_loss": 0.8441539406776428, | |
| "eval_runtime": 4.5501, | |
| "eval_samples_per_second": 101.097, | |
| "eval_steps_per_second": 3.297, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_accuracy": 0.8826086956521739, | |
| "eval_loss": 0.7064985632896423, | |
| "eval_runtime": 4.6867, | |
| "eval_samples_per_second": 98.149, | |
| "eval_steps_per_second": 3.201, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "grad_norm": 7.515535354614258, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.3317, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "eval_accuracy": 0.8891304347826087, | |
| "eval_loss": 0.6930130124092102, | |
| "eval_runtime": 4.6291, | |
| "eval_samples_per_second": 99.372, | |
| "eval_steps_per_second": 3.24, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "eval_accuracy": 0.8978260869565218, | |
| "eval_loss": 0.5629937648773193, | |
| "eval_runtime": 4.4716, | |
| "eval_samples_per_second": 102.872, | |
| "eval_steps_per_second": 3.355, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.26043701171875, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.1576, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8826086956521739, | |
| "eval_loss": 0.5882498025894165, | |
| "eval_runtime": 4.326, | |
| "eval_samples_per_second": 106.334, | |
| "eval_steps_per_second": 3.467, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "eval_accuracy": 0.908695652173913, | |
| "eval_loss": 0.5198363065719604, | |
| "eval_runtime": 4.262, | |
| "eval_samples_per_second": 107.93, | |
| "eval_steps_per_second": 3.519, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 9.87, | |
| "eval_accuracy": 0.9043478260869565, | |
| "eval_loss": 0.4424881935119629, | |
| "eval_runtime": 4.3408, | |
| "eval_samples_per_second": 105.972, | |
| "eval_steps_per_second": 3.456, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 10.67, | |
| "grad_norm": 2.913085460662842, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.0883, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "eval_accuracy": 0.8978260869565218, | |
| "eval_loss": 0.4727242887020111, | |
| "eval_runtime": 4.3328, | |
| "eval_samples_per_second": 106.168, | |
| "eval_steps_per_second": 3.462, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9021739130434783, | |
| "eval_loss": 0.4314464330673218, | |
| "eval_runtime": 4.6644, | |
| "eval_samples_per_second": 98.619, | |
| "eval_steps_per_second": 3.216, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "eval_accuracy": 0.9021739130434783, | |
| "eval_loss": 0.40110892057418823, | |
| "eval_runtime": 4.8419, | |
| "eval_samples_per_second": 95.004, | |
| "eval_steps_per_second": 3.098, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "grad_norm": 1.2048002481460571, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.051, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 13.87, | |
| "eval_accuracy": 0.9173913043478261, | |
| "eval_loss": 0.404487669467926, | |
| "eval_runtime": 4.5406, | |
| "eval_samples_per_second": 101.309, | |
| "eval_steps_per_second": 3.304, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 14.93, | |
| "eval_accuracy": 0.9108695652173913, | |
| "eval_loss": 0.3745279610157013, | |
| "eval_runtime": 4.2415, | |
| "eval_samples_per_second": 108.451, | |
| "eval_steps_per_second": 3.536, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 2.0225095748901367, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0415, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9152173913043479, | |
| "eval_loss": 0.3597088158130646, | |
| "eval_runtime": 4.2243, | |
| "eval_samples_per_second": 108.894, | |
| "eval_steps_per_second": 3.551, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 16.8, | |
| "eval_accuracy": 0.9065217391304348, | |
| "eval_loss": 0.40161189436912537, | |
| "eval_runtime": 4.2256, | |
| "eval_samples_per_second": 108.86, | |
| "eval_steps_per_second": 3.55, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 17.87, | |
| "eval_accuracy": 0.9152173913043479, | |
| "eval_loss": 0.3803761899471283, | |
| "eval_runtime": 4.2098, | |
| "eval_samples_per_second": 109.269, | |
| "eval_steps_per_second": 3.563, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 18.67, | |
| "grad_norm": 0.8873955607414246, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.0307, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 18.93, | |
| "eval_accuracy": 0.9217391304347826, | |
| "eval_loss": 0.351882666349411, | |
| "eval_runtime": 4.3177, | |
| "eval_samples_per_second": 106.538, | |
| "eval_steps_per_second": 3.474, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8934782608695652, | |
| "eval_loss": 0.4131200909614563, | |
| "eval_runtime": 4.523, | |
| "eval_samples_per_second": 101.703, | |
| "eval_steps_per_second": 3.316, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "eval_accuracy": 0.9, | |
| "eval_loss": 0.4047200679779053, | |
| "eval_runtime": 4.74, | |
| "eval_samples_per_second": 97.047, | |
| "eval_steps_per_second": 3.165, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 21.33, | |
| "grad_norm": 1.1633386611938477, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.0262, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 21.87, | |
| "eval_accuracy": 0.9173913043478261, | |
| "eval_loss": 0.34501612186431885, | |
| "eval_runtime": 4.8375, | |
| "eval_samples_per_second": 95.09, | |
| "eval_steps_per_second": 3.101, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 22.93, | |
| "eval_accuracy": 0.9108695652173913, | |
| "eval_loss": 0.36392879486083984, | |
| "eval_runtime": 4.5545, | |
| "eval_samples_per_second": 101.0, | |
| "eval_steps_per_second": 3.293, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 3.2603185176849365, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0208, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9043478260869565, | |
| "eval_loss": 0.38431569933891296, | |
| "eval_runtime": 4.371, | |
| "eval_samples_per_second": 105.239, | |
| "eval_steps_per_second": 3.432, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 24.8, | |
| "eval_accuracy": 0.8978260869565218, | |
| "eval_loss": 0.37965938448905945, | |
| "eval_runtime": 4.2698, | |
| "eval_samples_per_second": 107.734, | |
| "eval_steps_per_second": 3.513, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 25.87, | |
| "eval_accuracy": 0.9152173913043479, | |
| "eval_loss": 0.36596596240997314, | |
| "eval_runtime": 4.2147, | |
| "eval_samples_per_second": 109.141, | |
| "eval_steps_per_second": 3.559, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "grad_norm": 0.3663266599178314, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.0141, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 26.93, | |
| "eval_accuracy": 0.9152173913043479, | |
| "eval_loss": 0.34451282024383545, | |
| "eval_runtime": 4.4007, | |
| "eval_samples_per_second": 104.529, | |
| "eval_steps_per_second": 3.409, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9239130434782609, | |
| "eval_loss": 0.3131215572357178, | |
| "eval_runtime": 4.5955, | |
| "eval_samples_per_second": 100.098, | |
| "eval_steps_per_second": 3.264, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "eval_accuracy": 0.9195652173913044, | |
| "eval_loss": 0.30694660544395447, | |
| "eval_runtime": 4.7949, | |
| "eval_samples_per_second": 95.936, | |
| "eval_steps_per_second": 3.128, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 29.33, | |
| "grad_norm": 0.8458459973335266, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.0114, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 29.87, | |
| "eval_accuracy": 0.9195652173913044, | |
| "eval_loss": 0.30062630772590637, | |
| "eval_runtime": 4.6098, | |
| "eval_samples_per_second": 99.787, | |
| "eval_steps_per_second": 3.254, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 30.93, | |
| "eval_accuracy": 0.9239130434782609, | |
| "eval_loss": 0.30970004200935364, | |
| "eval_runtime": 4.5805, | |
| "eval_samples_per_second": 100.425, | |
| "eval_steps_per_second": 3.275, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.6790540814399719, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.014, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.9173913043478261, | |
| "eval_loss": 0.31208136677742004, | |
| "eval_runtime": 4.2755, | |
| "eval_samples_per_second": 107.59, | |
| "eval_steps_per_second": 3.508, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 32.8, | |
| "eval_accuracy": 0.9173913043478261, | |
| "eval_loss": 0.3241799473762512, | |
| "eval_runtime": 4.2702, | |
| "eval_samples_per_second": 107.722, | |
| "eval_steps_per_second": 3.513, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 33.87, | |
| "eval_accuracy": 0.9217391304347826, | |
| "eval_loss": 0.32906806468963623, | |
| "eval_runtime": 4.2635, | |
| "eval_samples_per_second": 107.894, | |
| "eval_steps_per_second": 3.518, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 34.67, | |
| "grad_norm": 0.5736334919929504, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.016, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 34.93, | |
| "eval_accuracy": 0.9217391304347826, | |
| "eval_loss": 0.31564629077911377, | |
| "eval_runtime": 4.3281, | |
| "eval_samples_per_second": 106.281, | |
| "eval_steps_per_second": 3.466, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9260869565217391, | |
| "eval_loss": 0.30808836221694946, | |
| "eval_runtime": 4.5545, | |
| "eval_samples_per_second": 100.999, | |
| "eval_steps_per_second": 3.293, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 36.8, | |
| "eval_accuracy": 0.9260869565217391, | |
| "eval_loss": 0.3083609640598297, | |
| "eval_runtime": 4.8866, | |
| "eval_samples_per_second": 94.135, | |
| "eval_steps_per_second": 3.07, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 37.33, | |
| "grad_norm": 1.5394645929336548, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.0114, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 37.87, | |
| "eval_accuracy": 0.9195652173913044, | |
| "eval_loss": 0.3148006796836853, | |
| "eval_runtime": 4.6556, | |
| "eval_samples_per_second": 98.806, | |
| "eval_steps_per_second": 3.222, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 38.93, | |
| "eval_accuracy": 0.9173913043478261, | |
| "eval_loss": 0.3191043436527252, | |
| "eval_runtime": 4.4147, | |
| "eval_samples_per_second": 104.198, | |
| "eval_steps_per_second": 3.398, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.860876202583313, | |
| "learning_rate": 0.0, | |
| "loss": 0.0091, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.9173913043478261, | |
| "eval_loss": 0.31911081075668335, | |
| "eval_runtime": 4.2782, | |
| "eval_samples_per_second": 107.523, | |
| "eval_steps_per_second": 3.506, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "step": 150, | |
| "total_flos": 4.585003492737024e+17, | |
| "train_loss": 0.09297615384062131, | |
| "train_runtime": 1186.4042, | |
| "train_samples_per_second": 19.386, | |
| "train_steps_per_second": 0.126 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 4.585003492737024e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |