| { | |
| "best_metric": 0.9122203098106713, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-azure-poc-img-classification/checkpoint-410", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 410, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24390243902439024, | |
| "grad_norm": 41.443572998046875, | |
| "learning_rate": 1.2195121951219513e-05, | |
| "loss": 1.6888, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4878048780487805, | |
| "grad_norm": 48.59061813354492, | |
| "learning_rate": 2.4390243902439026e-05, | |
| "loss": 1.1676, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7317073170731707, | |
| "grad_norm": 34.8450927734375, | |
| "learning_rate": 3.6585365853658535e-05, | |
| "loss": 0.8213, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.975609756097561, | |
| "grad_norm": 24.4506893157959, | |
| "learning_rate": 4.878048780487805e-05, | |
| "loss": 0.5888, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8347676419965576, | |
| "eval_loss": 0.4436015188694, | |
| "eval_runtime": 31.6183, | |
| "eval_samples_per_second": 18.375, | |
| "eval_steps_per_second": 0.601, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "grad_norm": 37.28171157836914, | |
| "learning_rate": 4.878048780487805e-05, | |
| "loss": 0.4412, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.4634146341463414, | |
| "grad_norm": 16.48870086669922, | |
| "learning_rate": 4.7425474254742554e-05, | |
| "loss": 0.3709, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.7073170731707317, | |
| "grad_norm": 28.970722198486328, | |
| "learning_rate": 4.607046070460705e-05, | |
| "loss": 0.3093, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.951219512195122, | |
| "grad_norm": 21.53656005859375, | |
| "learning_rate": 4.4715447154471546e-05, | |
| "loss": 0.3118, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8691910499139415, | |
| "eval_loss": 0.3027944266796112, | |
| "eval_runtime": 28.8996, | |
| "eval_samples_per_second": 20.104, | |
| "eval_steps_per_second": 0.657, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.1951219512195124, | |
| "grad_norm": 12.73299789428711, | |
| "learning_rate": 4.336043360433605e-05, | |
| "loss": 0.2855, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "grad_norm": 21.19304656982422, | |
| "learning_rate": 4.2005420054200545e-05, | |
| "loss": 0.2955, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.682926829268293, | |
| "grad_norm": 20.837289810180664, | |
| "learning_rate": 4.065040650406504e-05, | |
| "loss": 0.2377, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.926829268292683, | |
| "grad_norm": 11.421738624572754, | |
| "learning_rate": 3.9295392953929537e-05, | |
| "loss": 0.2284, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8795180722891566, | |
| "eval_loss": 0.28786736726760864, | |
| "eval_runtime": 27.9367, | |
| "eval_samples_per_second": 20.797, | |
| "eval_steps_per_second": 0.68, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 3.1707317073170733, | |
| "grad_norm": 23.18836784362793, | |
| "learning_rate": 3.794037940379404e-05, | |
| "loss": 0.2491, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.4146341463414633, | |
| "grad_norm": 14.66263198852539, | |
| "learning_rate": 3.6585365853658535e-05, | |
| "loss": 0.2057, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.658536585365854, | |
| "grad_norm": 20.51490592956543, | |
| "learning_rate": 3.523035230352303e-05, | |
| "loss": 0.2372, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.902439024390244, | |
| "grad_norm": 11.376835823059082, | |
| "learning_rate": 3.3875338753387534e-05, | |
| "loss": 0.203, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8950086058519794, | |
| "eval_loss": 0.2368348091840744, | |
| "eval_runtime": 29.5907, | |
| "eval_samples_per_second": 19.635, | |
| "eval_steps_per_second": 0.642, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 4.146341463414634, | |
| "grad_norm": 16.738306045532227, | |
| "learning_rate": 3.2520325203252037e-05, | |
| "loss": 0.2199, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.390243902439025, | |
| "grad_norm": 23.239999771118164, | |
| "learning_rate": 3.116531165311653e-05, | |
| "loss": 0.1974, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.634146341463414, | |
| "grad_norm": 20.330289840698242, | |
| "learning_rate": 2.9810298102981032e-05, | |
| "loss": 0.2177, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.878048780487805, | |
| "grad_norm": 10.759196281433105, | |
| "learning_rate": 2.8455284552845528e-05, | |
| "loss": 0.2254, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8984509466437177, | |
| "eval_loss": 0.2276138812303543, | |
| "eval_runtime": 29.2089, | |
| "eval_samples_per_second": 19.891, | |
| "eval_steps_per_second": 0.65, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 5.121951219512195, | |
| "grad_norm": 14.840445518493652, | |
| "learning_rate": 2.7100271002710027e-05, | |
| "loss": 0.1754, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.365853658536586, | |
| "grad_norm": 14.199987411499023, | |
| "learning_rate": 2.574525745257453e-05, | |
| "loss": 0.1711, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.609756097560975, | |
| "grad_norm": 13.013700485229492, | |
| "learning_rate": 2.4390243902439026e-05, | |
| "loss": 0.1842, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.853658536585366, | |
| "grad_norm": 16.866701126098633, | |
| "learning_rate": 2.3035230352303525e-05, | |
| "loss": 0.1976, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8967297762478486, | |
| "eval_loss": 0.2338663935661316, | |
| "eval_runtime": 28.264, | |
| "eval_samples_per_second": 20.556, | |
| "eval_steps_per_second": 0.672, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 6.097560975609756, | |
| "grad_norm": 12.873111724853516, | |
| "learning_rate": 2.1680216802168024e-05, | |
| "loss": 0.1773, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.341463414634147, | |
| "grad_norm": 24.094802856445312, | |
| "learning_rate": 2.032520325203252e-05, | |
| "loss": 0.1616, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.585365853658536, | |
| "grad_norm": 24.90325355529785, | |
| "learning_rate": 1.897018970189702e-05, | |
| "loss": 0.1792, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.829268292682927, | |
| "grad_norm": 18.315990447998047, | |
| "learning_rate": 1.7615176151761516e-05, | |
| "loss": 0.1603, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9036144578313253, | |
| "eval_loss": 0.2191040813922882, | |
| "eval_runtime": 29.3644, | |
| "eval_samples_per_second": 19.786, | |
| "eval_steps_per_second": 0.647, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 7.073170731707317, | |
| "grad_norm": 11.87928295135498, | |
| "learning_rate": 1.6260162601626018e-05, | |
| "loss": 0.1786, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.317073170731708, | |
| "grad_norm": 14.647104263305664, | |
| "learning_rate": 1.4905149051490516e-05, | |
| "loss": 0.1706, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.560975609756097, | |
| "grad_norm": 19.18857765197754, | |
| "learning_rate": 1.3550135501355014e-05, | |
| "loss": 0.1587, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.804878048780488, | |
| "grad_norm": 15.203424453735352, | |
| "learning_rate": 1.2195121951219513e-05, | |
| "loss": 0.1556, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9036144578313253, | |
| "eval_loss": 0.2248595505952835, | |
| "eval_runtime": 28.4808, | |
| "eval_samples_per_second": 20.4, | |
| "eval_steps_per_second": 0.667, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 8.048780487804878, | |
| "grad_norm": 9.790509223937988, | |
| "learning_rate": 1.0840108401084012e-05, | |
| "loss": 0.1462, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.292682926829269, | |
| "grad_norm": 12.157197952270508, | |
| "learning_rate": 9.48509485094851e-06, | |
| "loss": 0.1529, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.536585365853659, | |
| "grad_norm": 10.181126594543457, | |
| "learning_rate": 8.130081300813009e-06, | |
| "loss": 0.1524, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.78048780487805, | |
| "grad_norm": 7.913666725158691, | |
| "learning_rate": 6.775067750677507e-06, | |
| "loss": 0.1488, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9070567986230637, | |
| "eval_loss": 0.20175980031490326, | |
| "eval_runtime": 27.8089, | |
| "eval_samples_per_second": 20.893, | |
| "eval_steps_per_second": 0.683, | |
| "step": 369 | |
| }, | |
| { | |
| "epoch": 9.024390243902438, | |
| "grad_norm": 16.225345611572266, | |
| "learning_rate": 5.420054200542006e-06, | |
| "loss": 0.1574, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.268292682926829, | |
| "grad_norm": 22.599414825439453, | |
| "learning_rate": 4.0650406504065046e-06, | |
| "loss": 0.1444, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.512195121951219, | |
| "grad_norm": 14.447446823120117, | |
| "learning_rate": 2.710027100271003e-06, | |
| "loss": 0.1439, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 9.75609756097561, | |
| "grad_norm": 30.83609962463379, | |
| "learning_rate": 1.3550135501355015e-06, | |
| "loss": 0.1498, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 27.201889038085938, | |
| "learning_rate": 0.0, | |
| "loss": 0.158, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9122203098106713, | |
| "eval_loss": 0.21193988621234894, | |
| "eval_runtime": 28.3116, | |
| "eval_samples_per_second": 20.522, | |
| "eval_steps_per_second": 0.671, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 410, | |
| "total_flos": 1.2975913894293504e+18, | |
| "train_loss": 0.2908858444632553, | |
| "train_runtime": 2811.168, | |
| "train_samples_per_second": 18.569, | |
| "train_steps_per_second": 0.146 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 410, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 1.2975913894293504e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |