| { | |
| "best_metric": 0.9421487603305785, | |
| "best_model_checkpoint": "videomae-base-finetuned-good-gestureUnitsV2/checkpoint-119", | |
| "epoch": 9.1, | |
| "eval_steps": 500, | |
| "global_step": 170, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.058823529411764705, | |
| "grad_norm": 9.208045959472656, | |
| "learning_rate": 5.882352941176471e-06, | |
| "loss": 0.7018, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.2809917355371901, | |
| "eval_accuracy_gunit": 0.75, | |
| "eval_accuracy_nothing": 0.04938271604938271, | |
| "eval_loss": 0.7373607754707336, | |
| "eval_runtime": 15.9247, | |
| "eval_samples_per_second": 7.598, | |
| "eval_steps_per_second": 1.005, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.0176470588235293, | |
| "grad_norm": 6.387392520904541, | |
| "learning_rate": 9.803921568627451e-06, | |
| "loss": 0.6964, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.076470588235294, | |
| "grad_norm": 6.294075965881348, | |
| "learning_rate": 9.150326797385621e-06, | |
| "loss": 0.6812, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.32231404958677684, | |
| "eval_accuracy_gunit": 0.925, | |
| "eval_accuracy_nothing": 0.024691358024691357, | |
| "eval_loss": 0.7289611101150513, | |
| "eval_runtime": 15.5334, | |
| "eval_samples_per_second": 7.79, | |
| "eval_steps_per_second": 1.03, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0352941176470587, | |
| "grad_norm": 4.1350297927856445, | |
| "learning_rate": 8.496732026143791e-06, | |
| "loss": 0.6299, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0941176470588236, | |
| "grad_norm": 3.0245726108551025, | |
| "learning_rate": 7.84313725490196e-06, | |
| "loss": 0.606, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_accuracy": 0.5537190082644629, | |
| "eval_accuracy_gunit": 0.85, | |
| "eval_accuracy_nothing": 0.4074074074074074, | |
| "eval_loss": 0.6634026765823364, | |
| "eval_runtime": 19.667, | |
| "eval_samples_per_second": 6.152, | |
| "eval_steps_per_second": 0.814, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 3.052941176470588, | |
| "grad_norm": 5.276291847229004, | |
| "learning_rate": 7.189542483660131e-06, | |
| "loss": 0.604, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "eval_accuracy": 0.4297520661157025, | |
| "eval_accuracy_gunit": 0.95, | |
| "eval_accuracy_nothing": 0.1728395061728395, | |
| "eval_loss": 0.6623733639717102, | |
| "eval_runtime": 15.522, | |
| "eval_samples_per_second": 7.795, | |
| "eval_steps_per_second": 1.031, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.011764705882353, | |
| "grad_norm": 6.972124099731445, | |
| "learning_rate": 6.535947712418301e-06, | |
| "loss": 0.5941, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.070588235294117, | |
| "grad_norm": 5.121129989624023, | |
| "learning_rate": 5.882352941176471e-06, | |
| "loss": 0.5523, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "eval_accuracy": 0.8512396694214877, | |
| "eval_accuracy_gunit": 0.85, | |
| "eval_accuracy_nothing": 0.8518518518518519, | |
| "eval_loss": 0.550801157951355, | |
| "eval_runtime": 15.1821, | |
| "eval_samples_per_second": 7.97, | |
| "eval_steps_per_second": 1.054, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 5.029411764705882, | |
| "grad_norm": 6.430747985839844, | |
| "learning_rate": 5.2287581699346416e-06, | |
| "loss": 0.4295, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.088235294117647, | |
| "grad_norm": 6.107223033905029, | |
| "learning_rate": 4.5751633986928105e-06, | |
| "loss": 0.4609, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_accuracy": 0.859504132231405, | |
| "eval_accuracy_gunit": 0.875, | |
| "eval_accuracy_nothing": 0.8518518518518519, | |
| "eval_loss": 0.4713260531425476, | |
| "eval_runtime": 15.8201, | |
| "eval_samples_per_second": 7.649, | |
| "eval_steps_per_second": 1.011, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.047058823529412, | |
| "grad_norm": 7.5455217361450195, | |
| "learning_rate": 3.92156862745098e-06, | |
| "loss": 0.4126, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "eval_accuracy": 0.9421487603305785, | |
| "eval_accuracy_gunit": 0.85, | |
| "eval_accuracy_nothing": 0.9876543209876543, | |
| "eval_loss": 0.3118731379508972, | |
| "eval_runtime": 15.9904, | |
| "eval_samples_per_second": 7.567, | |
| "eval_steps_per_second": 1.001, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 7.0058823529411764, | |
| "grad_norm": 9.63264274597168, | |
| "learning_rate": 3.2679738562091506e-06, | |
| "loss": 0.325, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 7.064705882352941, | |
| "grad_norm": 6.023283004760742, | |
| "learning_rate": 2.6143790849673208e-06, | |
| "loss": 0.2594, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "eval_accuracy": 0.9421487603305785, | |
| "eval_accuracy_gunit": 0.85, | |
| "eval_accuracy_nothing": 0.9876543209876543, | |
| "eval_loss": 0.2562319338321686, | |
| "eval_runtime": 15.7832, | |
| "eval_samples_per_second": 7.666, | |
| "eval_steps_per_second": 1.014, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.023529411764706, | |
| "grad_norm": 13.239381790161133, | |
| "learning_rate": 1.96078431372549e-06, | |
| "loss": 0.3276, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 8.08235294117647, | |
| "grad_norm": 5.906073093414307, | |
| "learning_rate": 1.3071895424836604e-06, | |
| "loss": 0.1955, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "eval_accuracy": 0.9173553719008265, | |
| "eval_accuracy_gunit": 0.875, | |
| "eval_accuracy_nothing": 0.9382716049382716, | |
| "eval_loss": 0.2955465614795685, | |
| "eval_runtime": 16.2565, | |
| "eval_samples_per_second": 7.443, | |
| "eval_steps_per_second": 0.984, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 9.041176470588235, | |
| "grad_norm": 11.169438362121582, | |
| "learning_rate": 6.535947712418302e-07, | |
| "loss": 0.1869, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "grad_norm": 16.142396926879883, | |
| "learning_rate": 0.0, | |
| "loss": 0.2203, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "eval_accuracy": 0.9421487603305785, | |
| "eval_accuracy_gunit": 0.85, | |
| "eval_accuracy_nothing": 0.9876543209876543, | |
| "eval_loss": 0.23538190126419067, | |
| "eval_runtime": 15.9878, | |
| "eval_samples_per_second": 7.568, | |
| "eval_steps_per_second": 1.001, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "step": 170, | |
| "total_flos": 1.669726549892137e+18, | |
| "train_loss": 0.4637373012654922, | |
| "train_runtime": 478.7442, | |
| "train_samples_per_second": 2.841, | |
| "train_steps_per_second": 0.355 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "eval_accuracy": 0.8932038834951457, | |
| "eval_accuracy_gunit": 0.5416666666666666, | |
| "eval_accuracy_nothing": 1.0, | |
| "eval_loss": 0.35411152243614197, | |
| "eval_runtime": 14.7766, | |
| "eval_samples_per_second": 6.97, | |
| "eval_steps_per_second": 0.88, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "eval_accuracy": 0.8932038834951457, | |
| "eval_accuracy_gunit": 0.5416666666666666, | |
| "eval_accuracy_nothing": 1.0, | |
| "eval_loss": 0.35411134362220764, | |
| "eval_runtime": 13.8911, | |
| "eval_samples_per_second": 7.415, | |
| "eval_steps_per_second": 0.936, | |
| "step": 170 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 170, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "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": 1.669726549892137e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |