YOLOX Training Issues¶
Problems specific to YOLOX object detection.
CUDA Out of Memory with YOLOX¶
Solutions:
# 1. Reduce batch size
data = DetectionDataModule(batch_size=8, ...)
# 2. Use gradient accumulation for effective larger batch
trainer = AutoTrainer(accumulate_grad_batches=8, ...)
# 3. Use smaller model
model = YOLOXDetector(model_name="yolox-s", ...) # Instead of yolox-l
YOLOX Slow Training¶
Solutions:
# 1. Use mixed precision
trainer = AutoTrainer(precision="16-mixed", ...)
# 2. Reduce image size
data = DetectionDataModule(image_size=416, ...)
# 3. Reduce workers if CPU-bound
data = DetectionDataModule(num_workers=2, ...)
YOLOX Poor Performance¶
Solution: Use official settings:
model = YOLOXDetector(
model_name="yolox-s",
lr=0.01, # Official LR
optimizer="sgd", # SGD, not AdamW
scheduler="yolox", # YOLOX scheduler
total_epochs=300, # Full training
)
data = DetectionDataModule(batch_size=64, ...) # Proper batch size
Related Issues¶
- OOM Errors - Memory optimization
- Slow Training - Performance tips
- Convergence - Training issues