| { | |
| "best_metric": 0.15759548544883728, | |
| "best_model_checkpoint": "smart_tv_hand_gestures_image_detection/checkpoint-15190", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 15190, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.32916392363396973, | |
| "grad_norm": 1.4565895795822144, | |
| "learning_rate": 9.702774108322324e-07, | |
| "loss": 2.1185, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6583278472679395, | |
| "grad_norm": 1.595917820930481, | |
| "learning_rate": 9.372523117569352e-07, | |
| "loss": 1.8436, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.9874917709019092, | |
| "grad_norm": 2.2012250423431396, | |
| "learning_rate": 9.04227212681638e-07, | |
| "loss": 1.4844, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9256790123456791, | |
| "eval_loss": 1.3686660528182983, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 139.1921, | |
| "eval_samples_per_second": 87.289, | |
| "eval_steps_per_second": 10.913, | |
| "step": 1519 | |
| }, | |
| { | |
| "epoch": 1.316655694535879, | |
| "grad_norm": 1.8626253604888916, | |
| "learning_rate": 8.712021136063408e-07, | |
| "loss": 1.1357, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.6458196181698486, | |
| "grad_norm": 1.5412793159484863, | |
| "learning_rate": 8.381770145310435e-07, | |
| "loss": 0.8734, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.9749835418038182, | |
| "grad_norm": 1.9414352178573608, | |
| "learning_rate": 8.051519154557464e-07, | |
| "loss": 0.6898, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.98559670781893, | |
| "eval_loss": 0.6899833679199219, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 140.1792, | |
| "eval_samples_per_second": 86.675, | |
| "eval_steps_per_second": 10.836, | |
| "step": 3038 | |
| }, | |
| { | |
| "epoch": 2.3041474654377883, | |
| "grad_norm": 1.410414218902588, | |
| "learning_rate": 7.72126816380449e-07, | |
| "loss": 0.5598, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.633311389071758, | |
| "grad_norm": 1.2400822639465332, | |
| "learning_rate": 7.391017173051519e-07, | |
| "loss": 0.4651, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.9624753127057275, | |
| "grad_norm": 0.9760797023773193, | |
| "learning_rate": 7.060766182298547e-07, | |
| "loss": 0.3972, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9925925925925926, | |
| "eval_loss": 0.42788437008857727, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 138.4606, | |
| "eval_samples_per_second": 87.751, | |
| "eval_steps_per_second": 10.971, | |
| "step": 4557 | |
| }, | |
| { | |
| "epoch": 3.291639236339697, | |
| "grad_norm": 0.7729864120483398, | |
| "learning_rate": 6.730515191545574e-07, | |
| "loss": 0.3433, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.6208031599736668, | |
| "grad_norm": 0.7019414305686951, | |
| "learning_rate": 6.400264200792602e-07, | |
| "loss": 0.3052, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.9499670836076364, | |
| "grad_norm": 0.6397891044616699, | |
| "learning_rate": 6.070013210039629e-07, | |
| "loss": 0.2739, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9956378600823045, | |
| "eval_loss": 0.3098331689834595, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 140.6468, | |
| "eval_samples_per_second": 86.387, | |
| "eval_steps_per_second": 10.8, | |
| "step": 6076 | |
| }, | |
| { | |
| "epoch": 4.279131007241606, | |
| "grad_norm": 0.9204681515693665, | |
| "learning_rate": 5.739762219286658e-07, | |
| "loss": 0.2502, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.6082949308755765, | |
| "grad_norm": 0.8557950854301453, | |
| "learning_rate": 5.409511228533686e-07, | |
| "loss": 0.2293, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.937458854509546, | |
| "grad_norm": 0.5376387238502502, | |
| "learning_rate": 5.079260237780713e-07, | |
| "loss": 0.2135, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9967901234567901, | |
| "eval_loss": 0.2464514970779419, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 143.0601, | |
| "eval_samples_per_second": 84.929, | |
| "eval_steps_per_second": 10.618, | |
| "step": 7595 | |
| }, | |
| { | |
| "epoch": 5.266622778143516, | |
| "grad_norm": 2.1749770641326904, | |
| "learning_rate": 4.749009247027741e-07, | |
| "loss": 0.1991, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.595786701777485, | |
| "grad_norm": 0.4924432039260864, | |
| "learning_rate": 4.418758256274769e-07, | |
| "loss": 0.1887, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.924950625411455, | |
| "grad_norm": 0.44754961133003235, | |
| "learning_rate": 4.088507265521796e-07, | |
| "loss": 0.1787, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9973662551440329, | |
| "eval_loss": 0.20901748538017273, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 143.6019, | |
| "eval_samples_per_second": 84.609, | |
| "eval_steps_per_second": 10.578, | |
| "step": 9114 | |
| }, | |
| { | |
| "epoch": 6.254114549045425, | |
| "grad_norm": 1.0733314752578735, | |
| "learning_rate": 3.758256274768824e-07, | |
| "loss": 0.1706, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 6.583278472679394, | |
| "grad_norm": 0.45371586084365845, | |
| "learning_rate": 3.428005284015852e-07, | |
| "loss": 0.1637, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.912442396313364, | |
| "grad_norm": 0.3785471022129059, | |
| "learning_rate": 3.09775429326288e-07, | |
| "loss": 0.1571, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9988477366255144, | |
| "eval_loss": 0.1836911290884018, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 141.3146, | |
| "eval_samples_per_second": 85.978, | |
| "eval_steps_per_second": 10.749, | |
| "step": 10633 | |
| }, | |
| { | |
| "epoch": 7.2416063199473335, | |
| "grad_norm": 0.774605393409729, | |
| "learning_rate": 2.7675033025099076e-07, | |
| "loss": 0.1517, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.570770243581303, | |
| "grad_norm": 0.49195748567581177, | |
| "learning_rate": 2.437252311756935e-07, | |
| "loss": 0.1477, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 7.899934167215273, | |
| "grad_norm": 0.32700517773628235, | |
| "learning_rate": 2.107001321003963e-07, | |
| "loss": 0.1439, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9981069958847737, | |
| "eval_loss": 0.17038238048553467, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 142.0518, | |
| "eval_samples_per_second": 85.532, | |
| "eval_steps_per_second": 10.693, | |
| "step": 12152 | |
| }, | |
| { | |
| "epoch": 8.229098090849243, | |
| "grad_norm": 0.36188894510269165, | |
| "learning_rate": 1.7767503302509906e-07, | |
| "loss": 0.1402, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 8.558262014483212, | |
| "grad_norm": 0.38871052861213684, | |
| "learning_rate": 1.4464993394980185e-07, | |
| "loss": 0.1385, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.887425938117183, | |
| "grad_norm": 0.33406054973602295, | |
| "learning_rate": 1.1162483487450462e-07, | |
| "loss": 0.1359, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9986008230452675, | |
| "eval_loss": 0.16055729985237122, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 143.3485, | |
| "eval_samples_per_second": 84.758, | |
| "eval_steps_per_second": 10.597, | |
| "step": 13671 | |
| }, | |
| { | |
| "epoch": 9.216589861751151, | |
| "grad_norm": 0.32042938470840454, | |
| "learning_rate": 7.85997357992074e-08, | |
| "loss": 0.1341, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 9.545753785385122, | |
| "grad_norm": 0.29274484515190125, | |
| "learning_rate": 4.557463672391017e-08, | |
| "loss": 0.1333, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 9.874917709019092, | |
| "grad_norm": 0.3197895288467407, | |
| "learning_rate": 1.2549537648612946e-08, | |
| "loss": 0.1331, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9990123456790123, | |
| "eval_loss": 0.15759548544883728, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 142.0855, | |
| "eval_samples_per_second": 85.512, | |
| "eval_steps_per_second": 10.691, | |
| "step": 15190 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 15190, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 3.76634697648169e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |