| { | |
| "best_metric": 0.9675324675324676, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-70", | |
| "epoch": 30.0, | |
| "eval_steps": 500, | |
| "global_step": 150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6298701298701299, | |
| "eval_loss": 0.6615892648696899, | |
| "eval_runtime": 5.9895, | |
| "eval_samples_per_second": 25.712, | |
| "eval_steps_per_second": 0.835, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 6.212774753570557, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6583, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7597402597402597, | |
| "eval_loss": 0.5232259631156921, | |
| "eval_runtime": 5.5139, | |
| "eval_samples_per_second": 27.929, | |
| "eval_steps_per_second": 0.907, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7857142857142857, | |
| "eval_loss": 0.5042803287506104, | |
| "eval_runtime": 5.8851, | |
| "eval_samples_per_second": 26.168, | |
| "eval_steps_per_second": 0.85, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.873718738555908, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.3346, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8766233766233766, | |
| "eval_loss": 0.287881076335907, | |
| "eval_runtime": 5.8126, | |
| "eval_samples_per_second": 26.494, | |
| "eval_steps_per_second": 0.86, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.24241776764392853, | |
| "eval_runtime": 5.8885, | |
| "eval_samples_per_second": 26.152, | |
| "eval_steps_per_second": 0.849, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.3201093673706055, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.1544, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8896103896103896, | |
| "eval_loss": 0.22171010076999664, | |
| "eval_runtime": 5.7903, | |
| "eval_samples_per_second": 26.596, | |
| "eval_steps_per_second": 0.864, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.922077922077922, | |
| "eval_loss": 0.14658886194229126, | |
| "eval_runtime": 5.8797, | |
| "eval_samples_per_second": 26.192, | |
| "eval_steps_per_second": 0.85, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.792673587799072, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.088, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.948051948051948, | |
| "eval_loss": 0.12608817219734192, | |
| "eval_runtime": 5.723, | |
| "eval_samples_per_second": 26.909, | |
| "eval_steps_per_second": 0.874, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.922077922077922, | |
| "eval_loss": 0.16798314452171326, | |
| "eval_runtime": 5.7613, | |
| "eval_samples_per_second": 26.73, | |
| "eval_steps_per_second": 0.868, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 6.038075923919678, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.0977, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.935064935064935, | |
| "eval_loss": 0.14464019238948822, | |
| "eval_runtime": 5.8083, | |
| "eval_samples_per_second": 26.514, | |
| "eval_steps_per_second": 0.861, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.922077922077922, | |
| "eval_loss": 0.18122775852680206, | |
| "eval_runtime": 6.1597, | |
| "eval_samples_per_second": 25.001, | |
| "eval_steps_per_second": 0.812, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 4.068108558654785, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0719, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9285714285714286, | |
| "eval_loss": 0.17976221442222595, | |
| "eval_runtime": 6.0227, | |
| "eval_samples_per_second": 25.57, | |
| "eval_steps_per_second": 0.83, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.961038961038961, | |
| "eval_loss": 0.105579674243927, | |
| "eval_runtime": 5.8224, | |
| "eval_samples_per_second": 26.449, | |
| "eval_steps_per_second": 0.859, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 3.0755834579467773, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.0629, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9675324675324676, | |
| "eval_loss": 0.10734522342681885, | |
| "eval_runtime": 5.9253, | |
| "eval_samples_per_second": 25.99, | |
| "eval_steps_per_second": 0.844, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9545454545454546, | |
| "eval_loss": 0.11058878153562546, | |
| "eval_runtime": 5.9322, | |
| "eval_samples_per_second": 25.96, | |
| "eval_steps_per_second": 0.843, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.3675538301467896, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.0414, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9415584415584416, | |
| "eval_loss": 0.12863656878471375, | |
| "eval_runtime": 5.8097, | |
| "eval_samples_per_second": 26.508, | |
| "eval_steps_per_second": 0.861, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.961038961038961, | |
| "eval_loss": 0.07608745247125626, | |
| "eval_runtime": 5.9871, | |
| "eval_samples_per_second": 25.722, | |
| "eval_steps_per_second": 0.835, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.8409650325775146, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0397, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9675324675324676, | |
| "eval_loss": 0.07851818948984146, | |
| "eval_runtime": 5.9494, | |
| "eval_samples_per_second": 25.885, | |
| "eval_steps_per_second": 0.84, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9675324675324676, | |
| "eval_loss": 0.07459615916013718, | |
| "eval_runtime": 5.5847, | |
| "eval_samples_per_second": 27.575, | |
| "eval_steps_per_second": 0.895, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 40.9788932800293, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.0487, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9675324675324676, | |
| "eval_loss": 0.06842251867055893, | |
| "eval_runtime": 5.7506, | |
| "eval_samples_per_second": 26.78, | |
| "eval_steps_per_second": 0.869, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.961038961038961, | |
| "eval_loss": 0.06015988811850548, | |
| "eval_runtime": 5.6354, | |
| "eval_samples_per_second": 27.327, | |
| "eval_steps_per_second": 0.887, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 0.32140201330184937, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.0244, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.9675324675324676, | |
| "eval_loss": 0.055138688534498215, | |
| "eval_runtime": 5.6318, | |
| "eval_samples_per_second": 27.344, | |
| "eval_steps_per_second": 0.888, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.9675324675324676, | |
| "eval_loss": 0.06390709429979324, | |
| "eval_runtime": 5.9133, | |
| "eval_samples_per_second": 26.043, | |
| "eval_steps_per_second": 0.846, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.17090357840061188, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0214, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9675324675324676, | |
| "eval_loss": 0.05834332853555679, | |
| "eval_runtime": 5.6237, | |
| "eval_samples_per_second": 27.384, | |
| "eval_steps_per_second": 0.889, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.9675324675324676, | |
| "eval_loss": 0.06634216755628586, | |
| "eval_runtime": 5.782, | |
| "eval_samples_per_second": 26.634, | |
| "eval_steps_per_second": 0.865, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 15.199660301208496, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.0261, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.961038961038961, | |
| "eval_loss": 0.10064199566841125, | |
| "eval_runtime": 5.8966, | |
| "eval_samples_per_second": 26.117, | |
| "eval_steps_per_second": 0.848, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.9675324675324676, | |
| "eval_loss": 0.07111012190580368, | |
| "eval_runtime": 5.7198, | |
| "eval_samples_per_second": 26.924, | |
| "eval_steps_per_second": 0.874, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.7294249534606934, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.019, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9675324675324676, | |
| "eval_loss": 0.06293747574090958, | |
| "eval_runtime": 6.017, | |
| "eval_samples_per_second": 25.594, | |
| "eval_steps_per_second": 0.831, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.961038961038961, | |
| "eval_loss": 0.07278253883123398, | |
| "eval_runtime": 5.5819, | |
| "eval_samples_per_second": 27.589, | |
| "eval_steps_per_second": 0.896, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.9749196171760559, | |
| "learning_rate": 0.0, | |
| "loss": 0.0237, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.961038961038961, | |
| "eval_loss": 0.07468711584806442, | |
| "eval_runtime": 5.8121, | |
| "eval_samples_per_second": 26.496, | |
| "eval_steps_per_second": 0.86, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 150, | |
| "total_flos": 4.548645984411648e+17, | |
| "train_loss": 0.11414109120766322, | |
| "train_runtime": 869.5797, | |
| "train_samples_per_second": 21.045, | |
| "train_steps_per_second": 0.172 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 4.548645984411648e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |