It is recommended to use miniconda to manage the environment. After installing miniconda, run the following command to generate the environment and run it
conda create --name generate python=3.9 pillow pandas requests
conda activate generate
pip install python-dotenv
- modify configs in
src/config.py
- run
python src/get_table.py
, this will generate a table called ratio.csv, you can modify probability of feature occurrence in the ratio column. - run
python src/generate.py
to generate images. python src/final_check.py
, remove the duplicates and view the current probability distribution, which can be adjusted again.- (can skip)
python src/upload_mystery_box.py
push mystery box metadata to IPFS python src/upload.py
push data to IPFS
-
Generate a template: reads the parts folders, reads the files inside, checks the format and size, and generates a csv
-
Generate image: Fill in the probability of reading the csv and generate the image
-
Rename images: Support to delete some of the generated images and rename others
-
Upload image: upload image and metadata to ipfs and generate hash of tokenURL
-
Multiple folder support: Rare features can be placed in separate folders for generate
-
Upload mystery box data: generate mystery box metadata and upload
-
Generate images by attr.csv, it's use for update old images
You can use the Discussions to talk
0x49E53Fb3d5bf1532fEBAD88a1979E33A94844d1d