{ "best_metric": 0.5168846845626831, "best_model_checkpoint": "mushrooms_image_detection/checkpoint-19350", "epoch": 10.0, "eval_steps": 500, "global_step": 19350, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.26, "learning_rate": 9.766839378238344e-06, "loss": 4.0231, "step": 500 }, { "epoch": 0.52, "learning_rate": 9.50777202072539e-06, "loss": 3.1763, "step": 1000 }, { "epoch": 0.78, "learning_rate": 9.248704663212435e-06, "loss": 2.7159, "step": 1500 }, { "epoch": 1.0, "eval_accuracy": 0.6579440460037475, "eval_loss": 2.335684061050415, "eval_runtime": 255.9186, "eval_samples_per_second": 60.476, "eval_steps_per_second": 7.561, "step": 1935 }, { "epoch": 1.03, "learning_rate": 8.989637305699482e-06, "loss": 2.3776, "step": 2000 }, { "epoch": 1.29, "learning_rate": 8.730569948186529e-06, "loss": 2.0857, "step": 2500 }, { "epoch": 1.55, "learning_rate": 8.471502590673576e-06, "loss": 1.8722, "step": 3000 }, { "epoch": 1.81, "learning_rate": 8.212435233160623e-06, "loss": 1.6815, "step": 3500 }, { "epoch": 2.0, "eval_accuracy": 0.7631323899980617, "eval_loss": 1.5546201467514038, "eval_runtime": 273.4727, "eval_samples_per_second": 56.594, "eval_steps_per_second": 7.076, "step": 3870 }, { "epoch": 2.07, "learning_rate": 7.953367875647669e-06, "loss": 1.5331, "step": 4000 }, { "epoch": 2.33, "learning_rate": 7.694300518134716e-06, "loss": 1.3831, "step": 4500 }, { "epoch": 2.58, "learning_rate": 7.435233160621762e-06, "loss": 1.2778, "step": 5000 }, { "epoch": 2.84, "learning_rate": 7.176165803108809e-06, "loss": 1.1659, "step": 5500 }, { "epoch": 3.0, "eval_accuracy": 0.8074562253666732, "eval_loss": 1.152724027633667, "eval_runtime": 246.7059, "eval_samples_per_second": 62.735, "eval_steps_per_second": 7.843, "step": 5805 }, { "epoch": 3.1, "learning_rate": 6.917098445595856e-06, "loss": 1.0974, "step": 6000 }, { "epoch": 3.36, "learning_rate": 6.658031088082902e-06, "loss": 0.9875, "step": 6500 }, { "epoch": 3.62, "learning_rate": 6.398963730569949e-06, "loss": 0.9223, "step": 7000 }, { "epoch": 3.88, "learning_rate": 6.139896373056995e-06, "loss": 0.8759, "step": 7500 }, { "epoch": 4.0, "eval_accuracy": 0.8385345997286295, "eval_loss": 0.8977574110031128, "eval_runtime": 218.7108, "eval_samples_per_second": 70.765, "eval_steps_per_second": 8.847, "step": 7740 }, { "epoch": 4.13, "learning_rate": 5.880829015544042e-06, "loss": 0.809, "step": 8000 }, { "epoch": 4.39, "learning_rate": 5.621761658031088e-06, "loss": 0.7397, "step": 8500 }, { "epoch": 4.65, "learning_rate": 5.3626943005181356e-06, "loss": 0.7194, "step": 9000 }, { "epoch": 4.91, "learning_rate": 5.103626943005182e-06, "loss": 0.6707, "step": 9500 }, { "epoch": 5.0, "eval_accuracy": 0.8570136331330361, "eval_loss": 0.7434917092323303, "eval_runtime": 247.4876, "eval_samples_per_second": 62.536, "eval_steps_per_second": 7.819, "step": 9675 }, { "epoch": 5.17, "learning_rate": 4.844559585492228e-06, "loss": 0.6323, "step": 10000 }, { "epoch": 5.43, "learning_rate": 4.585492227979275e-06, "loss": 0.6029, "step": 10500 }, { "epoch": 5.68, "learning_rate": 4.326424870466322e-06, "loss": 0.558, "step": 11000 }, { "epoch": 5.94, "learning_rate": 4.067357512953368e-06, "loss": 0.5433, "step": 11500 }, { "epoch": 6.0, "eval_accuracy": 0.8712928862182594, "eval_loss": 0.6399914622306824, "eval_runtime": 221.8619, "eval_samples_per_second": 69.76, "eval_steps_per_second": 8.722, "step": 11610 }, { "epoch": 6.2, "learning_rate": 3.808290155440415e-06, "loss": 0.5238, "step": 12000 }, { "epoch": 6.46, "learning_rate": 3.549222797927461e-06, "loss": 0.4873, "step": 12500 }, { "epoch": 6.72, "learning_rate": 3.2901554404145083e-06, "loss": 0.4666, "step": 13000 }, { "epoch": 6.98, "learning_rate": 3.031088082901555e-06, "loss": 0.4732, "step": 13500 }, { "epoch": 7.0, "eval_accuracy": 0.8785294307682368, "eval_loss": 0.5811490416526794, "eval_runtime": 216.0872, "eval_samples_per_second": 71.624, "eval_steps_per_second": 8.955, "step": 13545 }, { "epoch": 7.24, "learning_rate": 2.772020725388601e-06, "loss": 0.4282, "step": 14000 }, { "epoch": 7.49, "learning_rate": 2.512953367875648e-06, "loss": 0.4227, "step": 14500 }, { "epoch": 7.75, "learning_rate": 2.2538860103626944e-06, "loss": 0.4143, "step": 15000 }, { "epoch": 8.0, "eval_accuracy": 0.8800801188860891, "eval_loss": 0.5482025742530823, "eval_runtime": 199.6943, "eval_samples_per_second": 77.503, "eval_steps_per_second": 9.69, "step": 15480 }, { "epoch": 8.01, "learning_rate": 1.994818652849741e-06, "loss": 0.405, "step": 15500 }, { "epoch": 8.27, "learning_rate": 1.7357512953367877e-06, "loss": 0.3876, "step": 16000 }, { "epoch": 8.53, "learning_rate": 1.4766839378238342e-06, "loss": 0.3792, "step": 16500 }, { "epoch": 8.79, "learning_rate": 1.217616580310881e-06, "loss": 0.3751, "step": 17000 }, { "epoch": 9.0, "eval_accuracy": 0.8842798992052724, "eval_loss": 0.525115430355072, "eval_runtime": 235.2802, "eval_samples_per_second": 65.781, "eval_steps_per_second": 8.224, "step": 17415 }, { "epoch": 9.04, "learning_rate": 9.585492227979275e-07, "loss": 0.371, "step": 17500 }, { "epoch": 9.3, "learning_rate": 6.994818652849742e-07, "loss": 0.3548, "step": 18000 }, { "epoch": 9.56, "learning_rate": 4.404145077720207e-07, "loss": 0.3594, "step": 18500 }, { "epoch": 9.82, "learning_rate": 1.8134715025906736e-07, "loss": 0.3606, "step": 19000 }, { "epoch": 10.0, "eval_accuracy": 0.8859598113329457, "eval_loss": 0.5168846845626831, "eval_runtime": 201.8296, "eval_samples_per_second": 76.683, "eval_steps_per_second": 9.587, "step": 19350 } ], "logging_steps": 500, "max_steps": 19350, "num_train_epochs": 10, "save_steps": 500, "total_flos": 4.801355152861962e+19, "trial_name": null, "trial_params": null }