learning_rate: 5e-05
train_batch_size: 16
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine_with_restarts
lr_scheduler_warmup_steps: 200
num_epochs: 4
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0152	0.9685	2000	0.0083	0.9822	0.9877	0.9849	0.9974
0.0069	1.9370	4000	0.0047	0.9888	0.9936	0.9912	0.9985
0.0029	2.9056	6000	0.0038	0.9911	0.9942	0.9927	0.9987
0.0012	3.8741	8000	0.0045	0.9912	0.9944	0.9928	0.9987
0.0008	4.0	8260	0.0045	0.9912	0.9944	0.9928	0.9987

jefftherover/pii-layout-synth-v15

pii-layout-synth-v15