| { | |
| "best_metric": 1.0, | |
| "best_model_checkpoint": "swinv2-base-patch4-window12to16-192to256-22kto1k-ft-finetuned-footulcer/checkpoint-325", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 325, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 14.772844314575195, | |
| "learning_rate": 1.5151515151515153e-05, | |
| "loss": 0.6758, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 17.389602661132812, | |
| "learning_rate": 3.0303030303030306e-05, | |
| "loss": 0.3891, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 33.26732635498047, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.4548, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 13.78071117401123, | |
| "learning_rate": 4.88013698630137e-05, | |
| "loss": 0.4219, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 13.317490577697754, | |
| "learning_rate": 4.708904109589041e-05, | |
| "loss": 0.5415, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 52.00563049316406, | |
| "learning_rate": 4.5376712328767126e-05, | |
| "loss": 0.425, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8793103448275862, | |
| "eval_loss": 0.2769334614276886, | |
| "eval_runtime": 2.9905, | |
| "eval_samples_per_second": 38.789, | |
| "eval_steps_per_second": 9.697, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 6.420164585113525, | |
| "learning_rate": 4.366438356164384e-05, | |
| "loss": 0.4124, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 6.303807258605957, | |
| "learning_rate": 4.195205479452055e-05, | |
| "loss": 0.3529, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 13.112977027893066, | |
| "learning_rate": 4.0239726027397265e-05, | |
| "loss": 0.3787, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 20.180587768554688, | |
| "learning_rate": 3.852739726027397e-05, | |
| "loss": 0.2384, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 19.817771911621094, | |
| "learning_rate": 3.6815068493150685e-05, | |
| "loss": 0.2643, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 14.556670188903809, | |
| "learning_rate": 3.51027397260274e-05, | |
| "loss": 0.3526, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 33.79677200317383, | |
| "learning_rate": 3.339041095890411e-05, | |
| "loss": 0.3182, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9827586206896551, | |
| "eval_loss": 0.054682787507772446, | |
| "eval_runtime": 2.2733, | |
| "eval_samples_per_second": 51.027, | |
| "eval_steps_per_second": 12.757, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 5.128925323486328, | |
| "learning_rate": 3.1678082191780824e-05, | |
| "loss": 0.2523, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 11.12570571899414, | |
| "learning_rate": 2.9965753424657534e-05, | |
| "loss": 0.1997, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 43.87520217895508, | |
| "learning_rate": 2.825342465753425e-05, | |
| "loss": 0.3346, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 7.066685199737549, | |
| "learning_rate": 2.654109589041096e-05, | |
| "loss": 0.3683, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "grad_norm": 11.333112716674805, | |
| "learning_rate": 2.4828767123287673e-05, | |
| "loss": 0.2765, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "grad_norm": 10.573657035827637, | |
| "learning_rate": 2.3116438356164386e-05, | |
| "loss": 0.2053, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9913793103448276, | |
| "eval_loss": 0.028553670272231102, | |
| "eval_runtime": 2.3027, | |
| "eval_samples_per_second": 50.376, | |
| "eval_steps_per_second": 12.594, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 9.268758773803711, | |
| "learning_rate": 2.1404109589041096e-05, | |
| "loss": 0.1977, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 62.42777633666992, | |
| "learning_rate": 1.969178082191781e-05, | |
| "loss": 0.3236, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "grad_norm": 23.722658157348633, | |
| "learning_rate": 1.7979452054794522e-05, | |
| "loss": 0.1492, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "grad_norm": 19.14108657836914, | |
| "learning_rate": 1.6267123287671235e-05, | |
| "loss": 0.2977, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "grad_norm": 6.952052116394043, | |
| "learning_rate": 1.4554794520547945e-05, | |
| "loss": 0.2568, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "grad_norm": 4.827333450317383, | |
| "learning_rate": 1.284246575342466e-05, | |
| "loss": 0.1743, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.0569567680358887, | |
| "learning_rate": 1.113013698630137e-05, | |
| "loss": 0.2892, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9913793103448276, | |
| "eval_loss": 0.01669893227517605, | |
| "eval_runtime": 2.2629, | |
| "eval_samples_per_second": 51.261, | |
| "eval_steps_per_second": 12.815, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "grad_norm": 16.4809627532959, | |
| "learning_rate": 9.417808219178083e-06, | |
| "loss": 0.1562, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "grad_norm": 5.131158828735352, | |
| "learning_rate": 7.705479452054794e-06, | |
| "loss": 0.1183, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "grad_norm": 8.427245140075684, | |
| "learning_rate": 5.993150684931507e-06, | |
| "loss": 0.0951, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "grad_norm": 1.835396647453308, | |
| "learning_rate": 4.280821917808219e-06, | |
| "loss": 0.1224, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "grad_norm": 2.6403584480285645, | |
| "learning_rate": 2.5684931506849316e-06, | |
| "loss": 0.1092, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "grad_norm": 41.65308380126953, | |
| "learning_rate": 8.561643835616439e-07, | |
| "loss": 0.1774, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0013297222321853042, | |
| "eval_runtime": 2.3045, | |
| "eval_samples_per_second": 50.337, | |
| "eval_steps_per_second": 12.584, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 325, | |
| "total_flos": 5.3252144691019776e+17, | |
| "train_loss": 0.2897487744918236, | |
| "train_runtime": 313.9626, | |
| "train_samples_per_second": 16.547, | |
| "train_steps_per_second": 1.035 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 325, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 5.3252144691019776e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |