| { | |
| "best_metric": 0.5859972834587097, | |
| "best_model_checkpoint": "facial_emotions_image_detection/checkpoint-15740", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 15740, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 9.7131931166348e-06, | |
| "loss": 1.5691, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 9.394518801784576e-06, | |
| "loss": 1.1566, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 9.075844486934355e-06, | |
| "loss": 1.005, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6603098927294399, | |
| "eval_loss": 1.0014679431915283, | |
| "eval_runtime": 113.4351, | |
| "eval_samples_per_second": 110.944, | |
| "eval_steps_per_second": 13.876, | |
| "step": 1574 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 8.757170172084131e-06, | |
| "loss": 0.9148, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 8.438495857233908e-06, | |
| "loss": 0.8522, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 8.119821542383684e-06, | |
| "loss": 0.8063, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7133889551052841, | |
| "eval_loss": 0.8330217003822327, | |
| "eval_runtime": 108.2894, | |
| "eval_samples_per_second": 116.216, | |
| "eval_steps_per_second": 14.535, | |
| "step": 3148 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 7.80114722753346e-06, | |
| "loss": 0.7566, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 7.482472912683239e-06, | |
| "loss": 0.7097, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 7.163798597833015e-06, | |
| "loss": 0.7016, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7423917361938817, | |
| "eval_loss": 0.7428602576255798, | |
| "eval_runtime": 107.858, | |
| "eval_samples_per_second": 116.681, | |
| "eval_steps_per_second": 14.593, | |
| "step": 4722 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 6.8451242829827925e-06, | |
| "loss": 0.6608, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 6.526449968132569e-06, | |
| "loss": 0.6235, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 6.207775653282346e-06, | |
| "loss": 0.6138, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7507350019864919, | |
| "eval_loss": 0.7060629725456238, | |
| "eval_runtime": 109.3785, | |
| "eval_samples_per_second": 115.059, | |
| "eval_steps_per_second": 14.39, | |
| "step": 6296 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 5.889101338432123e-06, | |
| "loss": 0.5723, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 5.5704270235819e-06, | |
| "loss": 0.5429, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 5.251752708731677e-06, | |
| "loss": 0.5403, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7726658720699245, | |
| "eval_loss": 0.6644242405891418, | |
| "eval_runtime": 108.6283, | |
| "eval_samples_per_second": 115.854, | |
| "eval_steps_per_second": 14.49, | |
| "step": 7870 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 4.933078393881454e-06, | |
| "loss": 0.5112, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 4.61440407903123e-06, | |
| "loss": 0.4647, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 4.295729764181008e-06, | |
| "loss": 0.4584, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7871275327771157, | |
| "eval_loss": 0.6358364820480347, | |
| "eval_runtime": 108.2696, | |
| "eval_samples_per_second": 116.238, | |
| "eval_steps_per_second": 14.538, | |
| "step": 9444 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 3.977055449330784e-06, | |
| "loss": 0.4567, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 3.6583811344805616e-06, | |
| "loss": 0.4134, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 3.339706819630338e-06, | |
| "loss": 0.3957, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 3.021032504780115e-06, | |
| "loss": 0.3991, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.800238379022646, | |
| "eval_loss": 0.6062497496604919, | |
| "eval_runtime": 108.4345, | |
| "eval_samples_per_second": 116.061, | |
| "eval_steps_per_second": 14.516, | |
| "step": 11018 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 2.702358189929892e-06, | |
| "loss": 0.3545, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 2.3836838750796687e-06, | |
| "loss": 0.3538, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.94, | |
| "learning_rate": 2.0650095602294456e-06, | |
| "loss": 0.3544, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8045292014302742, | |
| "eval_loss": 0.6008493900299072, | |
| "eval_runtime": 110.3834, | |
| "eval_samples_per_second": 114.012, | |
| "eval_steps_per_second": 14.259, | |
| "step": 12592 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "learning_rate": 1.7463352453792225e-06, | |
| "loss": 0.3183, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.58, | |
| "learning_rate": 1.4276609305289997e-06, | |
| "loss": 0.3218, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 1.1089866156787763e-06, | |
| "loss": 0.3184, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8116805721096544, | |
| "eval_loss": 0.590340256690979, | |
| "eval_runtime": 108.9419, | |
| "eval_samples_per_second": 115.52, | |
| "eval_steps_per_second": 14.448, | |
| "step": 14166 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "learning_rate": 7.903123008285534e-07, | |
| "loss": 0.3054, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 9.53, | |
| "learning_rate": 4.7163798597833015e-07, | |
| "loss": 0.2813, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 1.5296367112810708e-07, | |
| "loss": 0.2984, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8162097735399285, | |
| "eval_loss": 0.5859972834587097, | |
| "eval_runtime": 111.4622, | |
| "eval_samples_per_second": 112.908, | |
| "eval_steps_per_second": 14.121, | |
| "step": 15740 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 15740, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 3.900966581033497e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |