hchcsuim's picture
Training in progress, epoch 0
0b4a7f1 verified
{
"best_metric": 0.9983008958912574,
"best_model_checkpoint": "batch-size16_Celeb-DF_opencv-1FPS_faces-expand0-aligned_unaugmentation\\checkpoint-202",
"epoch": 0.9975308641975309,
"eval_steps": 500,
"global_step": 202,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04938271604938271,
"grad_norm": 2.2081010341644287,
"learning_rate": 2.380952380952381e-05,
"loss": 0.5797,
"step": 10
},
{
"epoch": 0.09876543209876543,
"grad_norm": 1.9404168128967285,
"learning_rate": 4.761904761904762e-05,
"loss": 0.4365,
"step": 20
},
{
"epoch": 0.14814814814814814,
"grad_norm": 4.0794453620910645,
"learning_rate": 4.751381215469613e-05,
"loss": 0.4103,
"step": 30
},
{
"epoch": 0.19753086419753085,
"grad_norm": 24.15789794921875,
"learning_rate": 4.475138121546962e-05,
"loss": 0.3143,
"step": 40
},
{
"epoch": 0.24691358024691357,
"grad_norm": 7.551469326019287,
"learning_rate": 4.1988950276243096e-05,
"loss": 0.185,
"step": 50
},
{
"epoch": 0.2962962962962963,
"grad_norm": 13.123407363891602,
"learning_rate": 3.9226519337016574e-05,
"loss": 0.1535,
"step": 60
},
{
"epoch": 0.345679012345679,
"grad_norm": 53.75169372558594,
"learning_rate": 3.646408839779006e-05,
"loss": 0.2063,
"step": 70
},
{
"epoch": 0.3950617283950617,
"grad_norm": 5.172009468078613,
"learning_rate": 3.370165745856354e-05,
"loss": 0.113,
"step": 80
},
{
"epoch": 0.4444444444444444,
"grad_norm": 6.01790714263916,
"learning_rate": 3.0939226519337016e-05,
"loss": 0.0534,
"step": 90
},
{
"epoch": 0.49382716049382713,
"grad_norm": 3.071531295776367,
"learning_rate": 2.8176795580110498e-05,
"loss": 0.063,
"step": 100
},
{
"epoch": 0.5432098765432098,
"grad_norm": 12.023387908935547,
"learning_rate": 2.541436464088398e-05,
"loss": 0.0826,
"step": 110
},
{
"epoch": 0.5925925925925926,
"grad_norm": 7.534938335418701,
"learning_rate": 2.265193370165746e-05,
"loss": 0.0631,
"step": 120
},
{
"epoch": 0.6419753086419753,
"grad_norm": 1.6553133726119995,
"learning_rate": 1.988950276243094e-05,
"loss": 0.0531,
"step": 130
},
{
"epoch": 0.691358024691358,
"grad_norm": 14.167679786682129,
"learning_rate": 1.712707182320442e-05,
"loss": 0.0559,
"step": 140
},
{
"epoch": 0.7407407407407407,
"grad_norm": 2.242429494857788,
"learning_rate": 1.4364640883977901e-05,
"loss": 0.0454,
"step": 150
},
{
"epoch": 0.7901234567901234,
"grad_norm": 4.238832950592041,
"learning_rate": 1.1602209944751381e-05,
"loss": 0.0468,
"step": 160
},
{
"epoch": 0.8395061728395061,
"grad_norm": 2.1989619731903076,
"learning_rate": 8.839779005524863e-06,
"loss": 0.0289,
"step": 170
},
{
"epoch": 0.8888888888888888,
"grad_norm": 13.035294532775879,
"learning_rate": 6.077348066298343e-06,
"loss": 0.0394,
"step": 180
},
{
"epoch": 0.9382716049382716,
"grad_norm": 9.14392375946045,
"learning_rate": 3.314917127071823e-06,
"loss": 0.0332,
"step": 190
},
{
"epoch": 0.9876543209876543,
"grad_norm": 2.9875118732452393,
"learning_rate": 5.524861878453039e-07,
"loss": 0.0172,
"step": 200
},
{
"epoch": 0.9975308641975309,
"eval_accuracy": 0.9983008958912574,
"eval_f1": 0.9989828000739782,
"eval_loss": 0.005187882110476494,
"eval_precision": 1.0,
"eval_recall": 0.9979676674364896,
"eval_roc_auc": 0.9999986075741439,
"eval_runtime": 224.8904,
"eval_samples_per_second": 57.575,
"eval_steps_per_second": 3.602,
"step": 202
},
{
"epoch": 0.9975308641975309,
"step": 202,
"total_flos": 3.213382256091464e+17,
"train_loss": 0.14779486708195494,
"train_runtime": 540.311,
"train_samples_per_second": 23.964,
"train_steps_per_second": 0.374
}
],
"logging_steps": 10,
"max_steps": 202,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"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": 3.213382256091464e+17,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}