$ pip install -r requirements.txt
There are nine training files in the folder, each file provide one training model. To set up the training files, users should change the training and testing folder listed in ImageFolder
.
# Running the Python script
# e.x. python main_swin_base_224.py
$ python <file>
To get the result, run:
$ python predict.py
Note that the users should change the weight path and save place of csv in
predict.py
.
Run the merge.ipynb
and fill up the the csv file's path. The generated file is called res.csv
.