Merge pull request #1 from hdr-bgnn/code-import

Adds initial code

.github/workflows/deploy-image.yml

@@ -0,0 +1,41 @@
+name: Create and publish a Docker image
+
+on:
+  release:
+    types: [published]
+
+env:
+  REGISTRY: ghcr.io
+  IMAGE_NAME: ${{ github.repository }}
+
+jobs:
+  build-and-push-image:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      packages: write
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v2
+
+      - name: Log in to the Container registry
+        uses: docker/login-action@f054a8b539a109f9f41c372932f1ae047eff08c9
+        with:
+          registry: ${{ env.REGISTRY }}
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Extract metadata (tags, labels) for Docker
+        id: meta
+        uses: docker/metadata-action@98669ae865ea3cffbcbaa878cf57c20bbf1c6c38
+        with:
+          images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
+
+      - name: Build and push Docker image
+        uses: docker/build-push-action@ad44023a93711e3deb337508980b4b5e9bcdc5dc
+        with:
+          context: .
+          push: true
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}

.github/workflows/run-tests.yml

@@ -0,0 +1,21 @@
+name: Run python tests
+
+on: [push]
+
+jobs:
+  build:
+
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ["3.7", "3.8", "3.9", "3.10"]
+
+    steps:
+      - uses: actions/checkout@v3
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v4
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Run python unit tests
+        run: |
+          python3 -m unittest discover .

Dockerfile

@@ -0,0 +1,11 @@
+FROM python:3.8.10-slim-buster
+LABEL "org.opencontainers.image.authors"="John Bradley <[email protected]>"
+LABEL "org.opencontainers.image.description"="Tool to reformat drexel metadata JSON files"
+
+WORKDIR /pipeline
+
+# ADD scripts in /pipeline to the PATH
+ENV PATH="/pipeline:${PATH}"
+
+COPY dm_formatter.py /pipeline
+CMD echo "dm_formatter.py input output"

README.md

@@ -1,2 +1,26 @@
 # drexel_metadata_formatter
-Reformats the metadata output file from drexel_metadata repo
+Reformats the metadata output file from the [drexel_metadata repo](https://github.com/hdr-bgnn/drexel_metadata/).
+
+## Requirements
+- [Python 3](https://www.python.org/)
+
+
+## Usage
+```
+usage: dm_formatter.py [-h] input output
+
+Convert metadata json file to BGNN minnow project format.
+
+positional arguments:
+  input       Path of input drexel_metadata format JSON metadata file.
+  output      Path of output BGNN minnow format JSON metadata file.
+
+optional arguments:
+  -h, --help  show this help message and exit
+ ```
+
+## Testing
+The unit tests can be run with the following command:
+```
+python3 -m unittest discover .
+```

dm_formatter.py

@@ -0,0 +1,85 @@
+#!/usr/bin/env python3
+# Convert drexel_metadata JSON file to BGNN minnow metadata JSON file
+import json
+import argparse
+
+
+def reformat_for_bgnn(result):
+    """
+    Reformat and reduce the size of the result dictionary. 
+    Collect only the data necessary for BGNN minnow project. The new format matches the 
+    BGNN_metadata version. Therefore some of the values not calculated in drexel version are by 
+    default set to "None". 
+
+    Parameters
+    ----------
+    result : dict
+        DESCRIPTION. output from gen_metadata()
+
+    Returns
+    -------
+    bgnn_result : dict
+        DESCRIPTION. {'base_name': xx, 'version':xx, 
+                       'fish': {'fish_num': xx,"bbox":xx, 'pixel_analysis':xx, 'rescale':xx, 
+                            'eye_bbox': xx, 'eye_center':xx , 'angle_degree': xx,
+                            'eye_direction':xx, 'foreground_mean':xx, 'background_mean':xx}, 
+                       'ruler': {'bbox':xx, 'scale':xx, 'unit':xx}}
+
+    """
+
+    name_base = list(result.keys())[0]
+    first_value = list(result.values())[0]
+
+    # Fish metadata
+    fish_num = first_value['fish_count']
+    fish_bbox = first_value['fish'][0]['bbox']
+    pixel_analysis = not first_value['fish'][0]['pixel_analysis_failed']
+
+    if first_value['fish'][0]['has_eye']:
+        eye_center = first_value['fish'][0]['eye_center']
+    else :
+        eye_center = "None"
+
+    eye_direction = first_value['fish'][0]['side']
+    foreground_mean = first_value['fish'][0]['foreground']['mean']
+    background_mean = first_value['fish'][0]['background']['mean']
+
+    dict_fish = {'fish_num': fish_num,"bbox": fish_bbox, 
+                 'pixel_analysis': pixel_analysis, 'rescale': "None", 
+                 'eye_bbox': "None", 'eye_center': eye_center , 'angle_degree': "None",
+                 'eye_direction': eye_direction, 'foreground_mean': round(foreground_mean,2), 
+                 'background_mean': round(background_mean,2)}
+
+    # Ruler metadata
+    ruler_bbox  = first_value['ruler_bbox'] if first_value['has_ruler'] else "None"
+    scale = round(first_value['scale'],2) if "scale" in first_value.keys() else "None"
+    unit = first_value['unit'] if "unit" in first_value.keys() else "None"
+
+    dict_ruler = {'bbox':ruler_bbox, 'scale':scale, 'unit':unit}
+
+    bgnn_result = {'base_name': name_base, 'version':"from drexel", 
+                   'fish': dict_fish, 'ruler': dict_ruler} 
+
+    return bgnn_result
+
+
+def argument_parser():
+    parser = argparse.ArgumentParser(description='Convert metadata json file to BGNN minnow project format.')
+    parser.add_argument('input', help='Path of input drexel_metadata format JSON metadata file.')
+    parser.add_argument('output', help='Path of output BGNN minnow format JSON metadata file.')
+    return parser
+
+
+def main():
+    parser = argument_parser()
+    args = parser.parse_args()
+    print(f"Converting {args.input} to BGNN minnows format {args.output}")
+    with open(args.input, 'r') as infile:
+        data = json.load(infile)
+        result = reformat_for_bgnn(data)
+        with open(args.output, 'w') as outfile:
+             outfile.write(json.dumps(result))
+
+
+if __name__ == '__main__':
+    main()

test_dm_formatter.py

@@ -0,0 +1,57 @@
+import unittest
+from dm_formatter import reformat_for_bgnn
+
+
+DM_INPUT_METADATA = {
+  "SOME-FISH-1234": {
+    "fish_count": 1,
+    "fish": [ {
+       "pixel_analysis_failed": False,
+       "bbox": [1,2,3,4],
+       "has_eye": True,
+       "eye_center": [50, 50],
+       "side": "left",
+       "foreground": {
+          "mean": 100.1111111,
+        },
+        "background": {
+          "mean": 200.1111111,
+        }
+       }
+    ],
+    "has_ruler": True,
+    "ruler_bbox": [
+      5,
+      6,
+      7,
+      8
+    ],
+  }
+}
+
+BGNN_EXPECTED_METADATA = {
+   'base_name': 'SOME-FISH-1234',
+   'fish': {'angle_degree': 'None',
+            'background_mean': 200.11,
+            'bbox': [1, 2, 3, 4],
+            'eye_bbox': 'None',
+            'eye_center': [50, 50],
+            'eye_direction': 'left',
+            'fish_num': 1,
+            'foreground_mean': 100.11,
+            'pixel_analysis': True,
+            'rescale': 'None'},
+   'ruler': {'bbox': [5, 6, 7, 8], 'scale': 'None', 'unit': 'None'},
+   'version': 'from drexel'
+}
+
+
+class TestFormatter(unittest.TestCase):
+    def test_reformat_for_bgnn(self):
+        result = reformat_for_bgnn(DM_INPUT_METADATA)
+        self.assertEqual(result, BGNN_EXPECTED_METADATA)
+
+
+
+if __name__ == '__main__':
+    unittest.main()