adaptive-style-transfer/evaluation at master · rajeevmsn/adaptive-style-transfer

Name		Name	Last commit message	Last commit date
parent directory ..
evaluation_data		evaluation_data
feature_extractor		feature_extractor
README.md		README.md
__init__.py		__init__.py
check_fc8_labels.py		check_fc8_labels.py
download_evaluation_data.py		download_evaluation_data.py
eval_deception_score.py		eval_deception_score.py
logger.py		logger.py
run_deception_score_vgg_16_wikiart.sh		run_deception_score_vgg_16_wikiart.sh

README.md

Run ./download_evaluation_data.py to download artist classification model.
Set results_dir variable in eval_deception_score.py:92 to point to the directory with stylized images. All images generated by one method must be in one directory. Image filenames must be in format "{content_name}_stylized_{artist_name}.jpg", for example: "Places366_val_00000510_stylized_vincent-van-gogh.jpg".
Run ./run_deception_score_vgg_16_wikiart.sh
Read results in the log file in ./logs directory.

Download validation sets from MSCOCO (val2017.zip) and Places365 (val_large.tar).
To compare with deception score reported in the paper run your stylization model on the content images listed in eval_paths_700_val.json.