| { | |
| "best_metric": 0.7092775390280552, | |
| "best_model_checkpoint": "levit_128.fb_dist_in1k-finetuned-stroke-binary/checkpoint-600", | |
| "epoch": 5.558139534883721, | |
| "eval_steps": 100, | |
| "global_step": 900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.31007751937984496, | |
| "grad_norm": 175.24969482421875, | |
| "learning_rate": 1.413793103448276e-06, | |
| "loss": 0.6951, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6201550387596899, | |
| "grad_norm": 146.73788452148438, | |
| "learning_rate": 3.137931034482759e-06, | |
| "loss": 0.695, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6201550387596899, | |
| "eval_accuracy": 0.5364088647670737, | |
| "eval_f1": 0.5420516079700721, | |
| "eval_loss": NaN, | |
| "eval_precision": 0.5540979766391916, | |
| "eval_recall": 0.5364088647670737, | |
| "eval_runtime": 6.9382, | |
| "eval_samples_per_second": 318.669, | |
| "eval_steps_per_second": 39.924, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9302325581395349, | |
| "grad_norm": 127.28129577636719, | |
| "learning_rate": 4.793103448275862e-06, | |
| "loss": 0.6942, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.235658914728682, | |
| "grad_norm": 751.1776123046875, | |
| "learning_rate": 6.482758620689655e-06, | |
| "loss": 0.6804, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.235658914728682, | |
| "eval_accuracy": 0.5798281320669381, | |
| "eval_f1": 0.5832968695954936, | |
| "eval_loss": NaN, | |
| "eval_precision": 0.6277845389250567, | |
| "eval_recall": 0.5798281320669381, | |
| "eval_runtime": 6.9012, | |
| "eval_samples_per_second": 320.38, | |
| "eval_steps_per_second": 40.138, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.5457364341085271, | |
| "grad_norm": 141.23553466796875, | |
| "learning_rate": 8.206896551724138e-06, | |
| "loss": 0.6896, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.8558139534883722, | |
| "grad_norm": 39.63947296142578, | |
| "learning_rate": 9.931034482758622e-06, | |
| "loss": 0.6821, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.8558139534883722, | |
| "eval_accuracy": 0.6232473993668024, | |
| "eval_f1": 0.6279682812403823, | |
| "eval_loss": NaN, | |
| "eval_precision": 0.6412845278993262, | |
| "eval_recall": 0.6232473993668024, | |
| "eval_runtime": 6.5774, | |
| "eval_samples_per_second": 336.153, | |
| "eval_steps_per_second": 42.114, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.1612403100775195, | |
| "grad_norm": 176.00396728515625, | |
| "learning_rate": 1.1655172413793106e-05, | |
| "loss": 0.6687, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.471317829457364, | |
| "grad_norm": 245.2242889404297, | |
| "learning_rate": 1.3379310344827587e-05, | |
| "loss": 0.6726, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.471317829457364, | |
| "eval_accuracy": 0.6671189507010402, | |
| "eval_f1": 0.6711377069586024, | |
| "eval_loss": NaN, | |
| "eval_precision": 0.6829477528090242, | |
| "eval_recall": 0.6671189507010402, | |
| "eval_runtime": 6.8482, | |
| "eval_samples_per_second": 322.86, | |
| "eval_steps_per_second": 40.449, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.781395348837209, | |
| "grad_norm": 17503.568359375, | |
| "learning_rate": 1.5068965517241381e-05, | |
| "loss": 0.6682, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.0868217054263565, | |
| "grad_norm": 69.87994384765625, | |
| "learning_rate": 1.6793103448275862e-05, | |
| "loss": 0.6546, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0868217054263565, | |
| "eval_accuracy": 0.70239710538218, | |
| "eval_f1": 0.7020627244507841, | |
| "eval_loss": NaN, | |
| "eval_precision": 0.7017573042623244, | |
| "eval_recall": 0.70239710538218, | |
| "eval_runtime": 6.8785, | |
| "eval_samples_per_second": 321.438, | |
| "eval_steps_per_second": 40.271, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.3968992248062015, | |
| "grad_norm": 101.73799133300781, | |
| "learning_rate": 1.8517241379310348e-05, | |
| "loss": 0.6508, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.7069767441860466, | |
| "grad_norm": 103.15814971923828, | |
| "learning_rate": 1.9999934702491228e-05, | |
| "loss": 0.647, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.7069767441860466, | |
| "eval_accuracy": 0.7064676616915423, | |
| "eval_f1": 0.7092775390280552, | |
| "eval_loss": NaN, | |
| "eval_precision": 0.7158812934752514, | |
| "eval_recall": 0.7064676616915423, | |
| "eval_runtime": 6.9098, | |
| "eval_samples_per_second": 319.98, | |
| "eval_steps_per_second": 40.088, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.0124031007751935, | |
| "grad_norm": 169.00445556640625, | |
| "learning_rate": 1.9994312394497127e-05, | |
| "loss": 0.6255, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.322480620155039, | |
| "grad_norm": 8452.361328125, | |
| "learning_rate": 1.997962682133918e-05, | |
| "loss": 0.6263, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.322480620155039, | |
| "eval_accuracy": 0.6956128448665763, | |
| "eval_f1": 0.6991433808448707, | |
| "eval_loss": NaN, | |
| "eval_precision": 0.709618757975013, | |
| "eval_recall": 0.6956128448665763, | |
| "eval_runtime": 6.9194, | |
| "eval_samples_per_second": 319.537, | |
| "eval_steps_per_second": 40.032, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.632558139534884, | |
| "grad_norm": 53.048866271972656, | |
| "learning_rate": 1.9955891300516223e-05, | |
| "loss": 0.6144, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.942635658914728, | |
| "grad_norm": 66.52362823486328, | |
| "learning_rate": 1.9923127356402373e-05, | |
| "loss": 0.6112, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.942635658914728, | |
| "eval_accuracy": 0.6766169154228856, | |
| "eval_f1": 0.6806907989098179, | |
| "eval_loss": NaN, | |
| "eval_precision": 0.7122760837780127, | |
| "eval_recall": 0.6766169154228856, | |
| "eval_runtime": 6.6311, | |
| "eval_samples_per_second": 333.428, | |
| "eval_steps_per_second": 41.773, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.248062015503876, | |
| "grad_norm": 39.8150634765625, | |
| "learning_rate": 1.9881364700727827e-05, | |
| "loss": 0.5889, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.558139534883721, | |
| "grad_norm": 156.32704162597656, | |
| "learning_rate": 1.9830641205634987e-05, | |
| "loss": 0.5704, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.558139534883721, | |
| "eval_accuracy": 0.6974219810040706, | |
| "eval_f1": 0.7013769772020675, | |
| "eval_loss": NaN, | |
| "eval_precision": 0.7297910740241028, | |
| "eval_recall": 0.6974219810040706, | |
| "eval_runtime": 6.7688, | |
| "eval_samples_per_second": 326.644, | |
| "eval_steps_per_second": 40.923, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.558139534883721, | |
| "step": 900, | |
| "total_flos": 2.1868569076064256e+17, | |
| "train_loss": 0.6519456799825033, | |
| "train_runtime": 297.561, | |
| "train_samples_per_second": 624.033, | |
| "train_steps_per_second": 19.478 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 5796, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 36, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.1868569076064256e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |