Packages included in requirements.txt
. Create a virtual environment and install:
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python3 -m venv venv
venv\Scripts\Activate.ps1
pip install -r requirements.txt
- Change into the directory
task1
:
cd ./task1
- Run
task1.py
:
python ./task1.py
This will output predicted angles, mean squared error, accuracy, and elapsed time to the console.
- Change into the directory
task2
:
cd ./task2
- Run
task2.py
:
python ./task2.py --test_directory_path ./test/images --train_directory_path ./train/png --sampling_levels 3 4
This will write prediction annotations and images to ./predict/annotations
and ./predict/images
respectively based on the templates in ./train/png
and test images in ./test/images
for sampling levels 3 and 4. You may wish to run sampling prediction in parallel in two separate shells for speed.
- Change into the directory
task2
:
cd ./task2
- Run
evaluate.py
:
python ./evaluate.py --pred_path_root ./predict/annotations/3 --gt_path_root ./test/annotations
This will output evaluation scores for the predictions in ./predict/annotations/3
against the ground truth values in ./test/annotations
.
- Change into the directory
task3
:
cd ./task3
- Run
task3.py
:
python ./task3.py --train_image_directory ./train_images/ --test_image_directory ./TestWithoutRotations/images/ --test_annotation_directory ./TestWithoutRotations/annotations/ --additional_test_image_directory ./Task3AdditionalTestDataset/images/ --additional_test_annotation_directory ./Task3AdditionalTestDataset/annotations/
This will output the TP, FP, FN and recall for each image. You can change the output type with the boolean SHOW and EVAL variables at the beginning of task3.py
.