Training of YoloXS Model on Coco dataset

Hi, I am currently reproducing the coco training on YoloXS model with line below:

python leondgarse/coco_train_script.py --det_header yolox.YOLOXS --data_name coco/2014 --batch_size 16

After my training using 30 epochs, I am getting poor result, as

# Show result
from keras_cv_attention_models.coco import data
data.show_image_with_bboxes(imm, bboxs, labels, confidences, num_classes=80)

Do I have anything configure wrongly? Or any suggestion could I change? Thanks!

[leondgarse]

Maybe caused by the rescale_mode. Try:

from keras_cv_attention_models import yolox

# rescale_mode from coco_train_script.py is `torch` by default. Original yolox is `raw`
# input_shape from coco_train_script.py is 256 by default
model = yolox.YOLOXS(pretrained='checkpoints/{mode_save_name}.h5', input_shape=(256, 256, 3), rescale_mode='torch')

# Run prediction
from keras_cv_attention_models import test_images
imm = test_images.dog_cat()
bboxs, lables, confidences = model.decode_predictions(model(model.preprocess_input(imm)))[0]

# Show result
from keras_cv_attention_models.coco import data
data.show_image_with_bboxes(imm, bboxs, lables, confidences, num_classes=80)

I actually still cannot claim it a reproduced, as my YOLOX_Tiny only reaches 31.8 using:

CUDA_VISIBLE_DEVICES='1' TF_GPU_ALLOCATOR='cuda_malloc_async' ./coco_train_script.py \
--det_header yolox.YOLOXTiny -p adamw --freeze_backbone_epochs 0 --lr_decay_steps 100 \
--positional_augment_methods tx -i 640 --rescale_mode tf

[leondgarse]

It’s from the anthor’s repo, also the BGR input format for the original YOLOX weights.