Merge pull request '#67_unittests_for_DTOs' (#69) from #67_unittests_for_DTOs into main

Reviewed-on: #69

.flake8

@@ -1,2 +0,0 @@
-[flake8]
-per-file-ignores = __init__.py:F401

.gitea/workflows/default.yaml

@@ -1,27 +0,0 @@
-name: Code Quality Pipeline
-run-name: ${{ gitea.actor }} is running the Code Quality Pipeline
-runs-on: ubuntu-latest
-on: push
-image: python:3.12
-jobs:
-  test:
-    name: Test
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout Code
-        uses: actions/checkout@v3
-      - name: Setup Environment
-        uses: https://github.com/actions/setup-python@v3
-        with:
-          python-verison: "3.12"
-          architecture: "x64"
-      - name: Install Packages
-        run: |
-          pip install poetry
-          poetry install
-      - name: PEP8 Check
-        run: |
-          poetry run flake8 ./src --benchmark
-      - name: Type Check
-        run: |
-          poetry run mypy ./src --disable-error-code=import-untyped

.gitea/workflows/publish.yaml

@@ -1,10 +1,8 @@
 name: CI Pipeline
 run-name: ${{ gitea.actor }} is running the CI Pipeline
-runs-on: ubuntu-latest
 on:
-  pull_request:
-    branches:
-      - main
+  release:
+    types: [published]
 jobs:
   test:
     name: Test
@@ -18,35 +16,60 @@ jobs:
           python-verison: "3.12"
           architecture: "x64"
       - name: Install Packages
+        env:
+          PIP_INDEX_URL: http://192.168.1.2:5001/index/
+          PIP_TRUSTED_HOST: 192.168.1.2
         run: |
           pip install poetry
           poetry install
       - name: PEP8 Check
         run: |
-          poetry run flake8 ./src --benchmark
+          poetry run flake8 . --benchmark
       - name: Type Check
         run: |
-          poetry run mypy ./src --disable-error-code=import-untyped
+          poetry run mypy .
   publish:
     name: Build and Publish
     runs-on: ubuntu-latest
     needs: [test]
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          -
+            dockerfile: ./Dockerfile.web_ui
+            image: ${{ vars.docker_repo_url }}/${{ gitea.repository }}/web_ui
+          -
+            dockerfile: ./Dockerfile.model
+            image: ${{ vars.docker_repo_url }}/${{ gitea.repository }}/model
     steps:
-      - name: Checkout Code
+      -
+        name: Checkout Code
         uses: actions/checkout@v3
-      - name: Set Environment Variables
+      -
+        name: Set Environment Variables
         run: |
           echo "sha_short=$(git rev-parse --short ${{ gitea.sha }} )" >> "$GITHUB_ENV"
-      - name: Show Environment Variables
+      -
+        name: Show Environment Variables
         run: |
           echo "DOCKER_REPO_URL: ${{ vars.docker_repo_url }}"
           echo "REPOSITORY: ${{ gitea.repository }}"
           echo "COMMIT_SHA: ${{ env.sha_short }}"
-      - name: Build and Push Image
-        uses: docker/build-push-action@v2
+      -
+        name: Extract metadata (tags, labels) for Docker
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ${{ matrix.image }}
+      -
+        name: Build and Push Image
+        uses: docker/build-push-action@v4
         with:
           context: .
+          file: ${{ matrix.dockerfile }}
           push: true
           tags: |
-            ${{ vars.docker_repo_url }}/${{ gitea.repository }}:${{ env.sha_short }}
-            ${{ vars.docker_repo_url }}/${{ gitea.repository }}:latest
+            ${{ matrix.image }}:${{ env.sha_short }}
+            ${{ matrix.image }}:latest
+          labels: ${{ steps.meta.outputs.labels }}

.gitea/workflows/pull.yaml

@@ -0,0 +1,50 @@
+name: Code Quality Pipeline
+on:
+  pull_request:
+    branches:
+      - main
+jobs:
+  job:
+    name: Check Code
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout Code
+        uses: actions/checkout@v4
+      - name: Setup Environment
+        uses: https://github.com/actions/setup-python@v5
+        with:
+          python-version: "3.12"
+          architecture: "x64"
+      - name: Setup poetry
+        env:
+          POETRY_VERSION: 2.1.1
+          POETRY_HOME: /opt/poetry
+          POETRY_NO_INTERACTION: 1
+          POETRY_NO_CACHE: 1
+        run: |
+          curl -sSL https://install.python-poetry.org | python3 -
+          export PATH=$POETRY_HOME/bin:$PATH
+          poetry --version
+      - name: Install Dependencies
+        env:
+          PIP_INDEX_URL: ${{ vars.PIP_INDEX_URL }}
+          PIP_TRUSTED_HOST: ${{ vars.PIP_TRUSTED_HOST }}
+        run: |
+          /opt/poetry/bin/poetry install
+      - name: PEP8 Check
+        run: |
+          /opt/poetry/bin/poetry run flake8 . --benchmark
+      - name: Type Check
+        run: |
+          /opt/poetry/bin/poetry run mypy .
+      - name: Pytest & Calculate Coverage
+        env:
+          MINIO_ENDPOINT: ${{ vars.MINIO_ENDPOINT }}
+          MINIO_ACCESS_KEY: ${{ vars.MINIO_ACCESS_KEY }}
+          MINIO_SECRET_KEY: ${{ secrets.MINIO_SECRET_KEY }}
+          MONGO_ENDPOINT: ${{ vars.MONGO_ENDPOINT }}
+        run: |
+          /opt/poetry/bin/poetry run coverage run -m pytest .
+      - name: Coverage Report
+        run: |
+          /opt/poetry/bin/poetry run coverage report -m

.gitea/workflows/release.yaml

@@ -0,0 +1,17 @@
+name: release-tag
+on:
+  workflow_dispatch:
+    inputs:
+      version:
+        description: 'The version number to deploy'
+        required: true
+jobs:
+  release-image:
+    runs-on: ubuntu-latest
+    steps:
+      -
+        run: echo "manual trigger executed"
+      -
+        run: echo "$MESSAGE"
+        env:
+          MESSAGE; ${{ github.event.inputs.version}}

.gitignore

@@ -177,3 +177,6 @@ ipython_config.py
 
 # Remove previous ipynb_checkpoints
 #   git rm -r .ipynb_checkpoints/
+
+# Logging data
+runs/*

.pre-commit-config.yaml

@@ -1,44 +1,62 @@
 repos:
 -   repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.5.0
+    rev: v5.0.0
     hooks:
     -   id: trailing-whitespace
     -   id: end-of-file-fixer
     -   id: check-yaml
+    -   id: check-added-large-files
     -   id: debug-statements
     -   id: double-quote-string-fixer
     -   id: name-tests-test
-    -   id: requirements-txt-fixer
 -   repo: https://github.com/asottile/setup-cfg-fmt
-    rev: v2.5.0
+    rev: v2.7.0
     hooks:
     -   id: setup-cfg-fmt
--   repo: https://github.com/asottile/reorder-python-imports
-    rev: v3.12.0
+-   repo: https://github.com/pre-commit/mirrors-isort
+    rev: v5.10.1
     hooks:
-    -   id: reorder-python-imports
-        exclude: ^(pre_commit/resources/|testing/resources/python3_hooks_repo/)
-        args: [--py39-plus, --add-import, 'from __future__ import annotations']
+      - id: isort
+        language_version: python3.10
+        args: [ --tc ]
 -   repo: https://github.com/asottile/add-trailing-comma
     rev: v3.1.0
     hooks:
     -   id: add-trailing-comma
 -   repo: https://github.com/asottile/pyupgrade
-    rev: v3.15.1
+    rev: v3.19.1
     hooks:
     -   id: pyupgrade
         args: [--py39-plus]
--   repo: https://github.com/hhatto/autopep8
-    rev: v2.0.4
-    hooks:
-    -   id: autopep8
 -   repo: https://github.com/PyCQA/flake8
     rev: 7.0.0
     hooks:
     -   id: flake8
+        entry: pflake8
+        additional_dependencies:
+          - "pyproject-flake8"
 -   repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.8.0
+    rev: v1.15.0
     hooks:
     -   id: mypy
-        additional_dependencies: [types-all]
         exclude: ^testing/resources/
+-   repo: https://github.com/psf/black
+    rev: 25.1.0
+    hooks:
+    - id: black
+      language_version: python3.12
+      args: [ --skip-string-normalization ]
+-   repo: https://github.com/myint/docformatter
+    rev: v1.5.0
+    hooks:
+      - id: docformatter
+        name: docformatter
+        description: 'Formats docstrings to follow PEP 257.'
+        entry: docformatter
+        language: python
+        types: [ python ]
+-     repo: https://github.com/jendrikseipp/vulture
+      rev: 'v2.14'
+      hooks:
+      - id: vulture
+        entry: vulture . --min-confidence 90 --exclude */.venv/*.py,*/tests/*.py

Dockerfile.model

@@ -0,0 +1,54 @@
+# build stage
+FROM python:3.12-slim-bookworm AS builder
+
+# set environment variables
+ENV PYTHONUNBUFFERED=1 \
+    PYTHONDONTWRITEBYTECODE=1 \
+    PIP_NO_CACHE_DIR=off \
+    PIP_DISABLE_PIP_VERSION_CHECK=ON \
+    PIP_DEFAULT_TIMEOUT=100 \
+    DEBIAN_FRONTEND=noninteractive \
+    POETRY_HOME=/etc/poetry \
+    POETRY_VERSION=1.7.1 \
+    POETRY_VIRTUALENVS_IN_PROJECT=1 \
+    POETRY_VIRTUALENVS_CREATE=1 \
+    POETRY_NO_INTERACTION=1 \
+    POETRY_CACHE_DIR=/tmp/poetry_cache \
+    APP_HOME=/home/app
+
+# update system
+RUN apt-get update \
+    && apt-get install -y --no-install-recommends \
+    build-essential \
+    curl \
+    && apt-get clean
+
+# install poetry
+RUN curl -sSL https://install.python-poetry.org | python3 -
+ENV PATH="${POETRY_HOME}/bin:$PATH"
+
+# install runtime dependencies
+WORKDIR ${APP_HOME}
+COPY ./poetry.lock ./pyproject.toml ./
+RUN --mount=type=cache,target=${POETRY_CACHE_DIR} poetry install \
+    --with shared,model \
+    --no-root
+
+# final stage
+FROM python:3.12-slim-bookworm
+
+# copy virtualenv made by poetry
+ENV APP_HOME=/home/app \
+    VIRTUAL_ENV=/home/app/.venv
+COPY --from=builder ${VIRTUAL_ENV} ${VIRTUAL_ENV}
+ENV PATH="${VIRTUAL_ENV}/bin:${PATH}"
+
+# create home directory and app user
+RUN mkdir -p $APP_HOME
+
+# add code while changing ownership
+WORKDIR $APP_HOME
+COPY ./shared ./shared
+COPY ./model/src ./
+
+ENTRYPOINT [ "python", "main.py" ]

Dockerfile.web_ui

@@ -0,0 +1,59 @@
+# build stage
+FROM python:3.12-slim-bookworm AS builder
+
+# set environment variables
+ENV PYTHONUNBUFFERED=1 \
+    PYTHONDONTWRITEBYTECODE=1 \
+    PIP_NO_CACHE_DIR=off \
+    PIP_DISABLE_PIP_VERSION_CHECK=ON \
+    PIP_DEFAULT_TIMEOUT=100 \
+    DEBIAN_FRONTEND=noninteractive \
+    POETRY_HOME=/etc/poetry \
+    POETRY_VERSION=1.7.1 \
+    POETRY_VIRTUALENVS_IN_PROJECT=1 \
+    POETRY_VIRTUALENVS_CREATE=1 \
+    POETRY_NO_INTERACTION=1 \
+    POETRY_CACHE_DIR=/tmp/poetry_cache \
+    APP_HOME=/home/app
+
+# update system
+RUN apt-get update \
+    && apt-get install -y --no-install-recommends \
+    build-essential \
+    curl \
+    && apt-get clean
+
+# install poetry
+RUN curl -sSL https://install.python-poetry.org | python3 -
+ENV PATH="${POETRY_HOME}/bin:$PATH"
+
+# install runtime dependencies
+WORKDIR ${APP_HOME}
+COPY ./poetry.lock ./pyproject.toml ./
+RUN --mount=type=cache,target=${POETRY_CACHE_DIR} poetry install \
+    --with shared,web_ui \
+    --no-root
+
+# final stage
+FROM python:3.12-slim-bookworm
+
+# copy virtualenv made by poetry
+ENV APP_HOME=/home/app \
+    VIRTUAL_ENV=/home/app/.venv
+COPY --from=builder ${VIRTUAL_ENV} ${VIRTUAL_ENV}
+ENV PATH="${VIRTUAL_ENV}/bin:${PATH}"
+
+# create home directory and app user
+RUN mkdir -p /home/app && \
+    addgroup --system app && \
+    adduser --system --group app
+
+# add code while changing ownership
+WORKDIR $APP_HOME
+COPY --chown=app:app ./shared ./shared
+COPY --chown=app:app ./web_ui/src ./src
+
+# change to the app user
+USER app
+
+ENTRYPOINT [ "gunicorn", "src.main:server", "-b", "0.0.0.0:8050" ]

core/database/__init__.py

@@ -1,13 +0,0 @@
-from __future__ import annotations
-
-from .classes import ModelOutputs
-from .classes import NoDocumentFoundException
-from .classes import VisualCommunication
-from .database import connect
-from .utils import get_visual_communication
-from .utils import list_names
-from .utils import total_annotated
-from .utils import total_documents
-from .utils import upsert_annotations
-from .utils import upsert_predictions
-from .utils import upsert_visual_communication

core/database/classes.py

@@ -1,165 +0,0 @@
-from __future__ import annotations
-
-import logging
-from base64 import b64decode
-from base64 import b64encode
-from io import BytesIO
-from pathlib import Path
-
-from PIL import Image
-from pydantic import BaseModel
-from pydantic import field_serializer
-from pydantic import field_validator
-
-from src.model_experiential import (
-    VisualSyntaxModelOutput,
-)
-from src.model_interpersonal import AngleModelOutput
-from src.model_interpersonal import ContactModelOutput
-from src.model_interpersonal import DistanceModelOutput
-from src.model_interpersonal import ModalityColorModelOutput
-from src.model_interpersonal import ModalityDepthModelOutput
-from src.model_interpersonal import ModalityLightingModelOutput
-from src.model_interpersonal import PointOfViewModelOutput
-from src.model_textual import FramingModelOutput
-from src.model_textual import InformationValueModelOutput
-from src.model_textual import SalienceModelOutput
-
-
-class NoDocumentFoundException(Exception):
-    pass
-
-
-class ModelOutputs(BaseModel):
-    visual_syntax: VisualSyntaxModelOutput
-    contact: ContactModelOutput
-    angle: AngleModelOutput
-    point_of_view: PointOfViewModelOutput
-    distance: DistanceModelOutput
-    modality_lighting: ModalityLightingModelOutput
-    modality_color: ModalityColorModelOutput
-    modality_depth: ModalityDepthModelOutput
-    information_value: InformationValueModelOutput
-    framing: FramingModelOutput
-    salience: SalienceModelOutput
-
-    @classmethod
-    def list_fields(cls) -> list[str]:
-        """List options that are stored as attributes."""
-        return list(cls.model_fields.keys())
-
-    @classmethod
-    def from_random(cls) -> ModelOutputs:
-        """Instantiate with random numbers."""
-        kwargs = {
-            field: field_info.annotation.from_random()  # type: ignore
-            for field, field_info
-            in cls.model_fields.items()
-        }
-        return cls(**kwargs)
-
-    @classmethod
-    def from_annotations(
-        cls,
-        visual_syntax: str,
-        contact: str,
-        angle: str,
-        point_of_view: str,
-        distance: str,
-        modality_lighting: str,
-        modality_color: str,
-        modality_depth: str,
-        information_value: str,
-        framing: str,
-        salience: str,
-    ) -> ModelOutputs:
-        """Instantiate from annotation."""
-        kwargs = {
-            'visual_syntax': VisualSyntaxModelOutput
-            .from_choice(visual_syntax),
-            'contact': ContactModelOutput
-            .from_choice(contact),
-            'angle': AngleModelOutput
-            .from_choice(angle),
-            'point_of_view': PointOfViewModelOutput
-            .from_choice(point_of_view),
-            'distance': DistanceModelOutput
-            .from_choice(distance),
-            'modality_lighting': ModalityLightingModelOutput
-            .from_choice(modality_lighting),
-            'modality_color': ModalityColorModelOutput
-            .from_choice(modality_color),
-            'modality_depth': ModalityDepthModelOutput
-            .from_choice(modality_depth),
-            'information_value': InformationValueModelOutput
-            .from_choice(information_value),
-            'framing': FramingModelOutput
-            .from_choice(framing),
-            'salience': SalienceModelOutput
-            .from_choice(salience),
-        }
-        return cls(**kwargs)
-
-
-class VisualCommunication(BaseModel):
-    name: str
-    image: Image.Image
-    annotation: ModelOutputs | None = None
-    prediction: ModelOutputs | None = None
-
-    class Config:
-        arbitrary_types_allowed = True
-
-    @classmethod
-    def classname(cls) -> str:
-        """Return classname."""
-        return cls.__name__
-
-    @classmethod
-    def from_file(cls, path: Path) -> VisualCommunication:
-        """Instantiate from file."""
-        name = path.stem
-        image = Image.open(path)
-        image.load()
-        return VisualCommunication(name=name, image=image)
-
-    @classmethod
-    def decode_image(cls, content: str) -> Image.Image:
-        """Decode image."""
-        _, content_data = content.split(',')
-        return Image.open(BytesIO(b64decode(content_data)))
-
-    @field_serializer('image')
-    def serialize_image(image: Image.Image) -> bytes:   # type: ignore
-        buffer = BytesIO()
-        image.save(buffer, format='JPEG')
-        return buffer.getvalue()
-
-    @field_validator('image', mode='before')
-    @classmethod
-    def convert_to_image(
-        cls,
-        image: Image.Image | BytesIO | bytes,
-    ) -> Image.Image:
-        if isinstance(image, bytes):
-            image = BytesIO(image)
-        if isinstance(image, BytesIO):
-            image = Image.open(image)
-        return image
-
-    def __repr__(self) -> str:
-        return f"{self.classname()}(name='{self.name}')"
-
-    def webencoded_image(self) -> str:
-        """Convert image to be displayed on webpage."""
-        # convert images to bytes string
-        buffer = BytesIO()
-        self.image.save(buffer, format='png')
-        img_enc = b64encode(buffer.getvalue()).decode('utf-8')
-        return f"data:image/png;base64, {img_enc}"
-
-    def generate_random_prediction(self, force: bool = False) -> None:
-        """Generate random prediction values."""
-        if not force and self.prediction is not None:
-            logging.warning('set force=True to overwrite existing values.')
-        self.prediction = ModelOutputs.from_random()

core/database/database.py

@@ -1,27 +0,0 @@
-from __future__ import annotations
-
-import logging
-import os
-
-from dotenv import load_dotenv
-from pymongo import MongoClient
-
-
-def connect():
-    """Connect to MongoDB."""
-    # load env vars
-    load_dotenv()
-    necessary_env_vars = [
-        'MONGO_HOST',
-        'MONGO_DB',
-        'MONGO_COLLECTION',
-    ]
-    for env_var in necessary_env_vars:
-        assert env_var in os.environ, f"{env_var} not found"
-    # connect to database
-    client = MongoClient(os.getenv('MONGO_HOST'))
-    db = client[os.getenv('MONGO_DB')]
-    collection = db[os.getenv('MONGO_COLLECTION')]
-    collection.create_index('name', unique=True)
-    logging.info('connected to database')
-    return collection, db, client

core/database/utils.py

@@ -1,158 +0,0 @@
-from __future__ import annotations
-
-import logging
-
-from pymongo.collection import Collection
-
-from .classes import ModelOutputs
-from .classes import NoDocumentFoundException
-from .classes import VisualCommunication
-
-
-def count_documents(
-    collection: Collection,
-    has_annotation: bool = False,
-) -> int:
-    """
-    Get the total number of documents
-    in database that matches the filters.
-    """
-    assert isinstance(collection, Collection)
-    assert isinstance(has_annotation, bool)
-    # build query
-    query = {}
-    if has_annotation:
-        query['annotation'] = {'$ne': None}
-    return collection.count_documents(filter=query)
-
-
-def list_names(
-    collection: Collection,
-    has_annotation: bool = True,
-) -> list[str]:
-    """List the names of entries that match the filters."""
-    assert isinstance(collection, Collection)
-    assert isinstance(has_annotation, bool)
-    # build query
-    query = {}
-    if has_annotation:
-        query['annotation'] = {'$ne': None}
-    res_list = collection.find(
-        filter=query,
-        projection={
-            '_id': False,
-            'name': True,
-        },
-    )
-    return list(res_list)
-
-
-def total_documents(
-    collection: Collection,
-) -> int:
-    """Get total number of documents in database."""
-    return collection.count_documents(filter={})
-
-
-def total_annotated(
-    collection: Collection,
-) -> int:
-    """Get total number of annotated documents in database."""
-    query = {
-        'annotation': {
-            '$ne': None,
-        },
-    }
-    return collection.count_documents(filter=query)
-
-
-def get_visual_communication(
-    collection: Collection,
-    with_annotation: bool = False,
-) -> VisualCommunication:
-    """Get a random visual communication from the database."""
-    query = {}
-    if with_annotation:
-        query['annotation'] = {'$ne': None}
-    else:
-        query['annotation'] = {'$eq': None}
-    data = collection.aggregate([
-        {
-            '$match': query,     # find using filters
-        },
-        {
-            '$sample': {
-                'size': 1,   # get one random
-            },
-        },
-    ])
-    data_list = list(data)   # read data from cursor object
-    if len(data_list) == 0:
-        logging.error('failed getting visual communication')
-        raise NoDocumentFoundException()
-    logging.info('finished')
-    return VisualCommunication.model_validate(data_list[0])
-
-
-def upsert_predictions(
-    collection: Collection,
-    vis_com_name: str,
-    predictions: ModelOutputs,
-) -> None:
-    """Upsert prediction data in the database."""
-    query = {
-        'name': vis_com_name,
-    }
-    update = {
-        '$set': {
-            'prediction': predictions.model_dump(),
-        },
-    }
-    res = collection.update_one(
-        filter=query,
-        update=update,
-        upsert=True,
-    )
-    logging.debug('upserted document: %s', res)
-    logging.info('finished')
-
-
-def upsert_annotations(
-    collection: Collection,
-    vis_com_name: str,
-    annotations: ModelOutputs,
-) -> None:
-    """Upserts annotation data in the database."""
-    query = {
-        'name': vis_com_name,
-    }
-    update = {
-        '$set': {
-            'annotation': annotations.model_dump(),
-        },
-    }
-    res = collection.update_one(
-        filter=query,
-        update=update,
-        upsert=True,
-    )
-    logging.info('upserted document: %s', res)
-    logging.info('finished')
-
-
-def upsert_visual_communication(
-    collection: Collection,
-    visual_communication_list: list[VisualCommunication],
-) -> bool:
-    """
-    Upsert VisualCommunication object in the database.
-    Returns bool stating success.
-    """
-    response = collection.insert_many(
-        [
-            vis_com.model_dump()
-            for vis_com
-            in visual_communication_list
-        ],
-    )
-    return response.acknowledged

docker-compose.local.yml

@@ -1,11 +1,10 @@
-version: '3.7'
 services:
   app:
     image: visual_critical_discourse_analysis:dev
     container_name: visual_critical_discourse_analysis
     build:
       context: .
-      dockerfile: Dockerfile
+      dockerfile: ./web_ui/Dockerfile
     env_file:
       - local.env
     environment:
@@ -41,4 +40,4 @@ services:
 
 networks:
   backend:
-    driver: bridge
+    driver: bridge

docker-compose.server.yml

@@ -1,11 +1,10 @@
-version: '3.7'
 services:
   app:
     image: visual_critical_discourse_analysis:dev
     container_name: visual_critical_discourse_analysis
     build:
       context: .
-      dockerfile: Dockerfile
+      dockerfile: ./web_ui/Dockerfile
     env_file:
       - server.env
     ports:

image_download/get_images.py

@@ -6,10 +6,9 @@ from pathlib import Path
 
 import pandas as pd
 import requests
+from classes import Instagram
 from retry import retry
 
-from image_download.classes import Instagram
-
 
 def get_sources() -> pd.DataFrame:
     """Get sources dateframe."""

misc/image_download.py

@@ -3,10 +3,11 @@ from __future__ import annotations
 import os
 from pathlib import Path
 
-from database import connect
-from database import VisualCommunication
 from dotenv import load_dotenv
 
+from shared.docstore import connect_mongodb
+from shared.docstore.classes import VisualCommunication
+
 if __name__ == '__main__':
     # prepare env vars
     env_path = Path(__file__).parent.parent / 'local.env'
@@ -14,7 +15,7 @@ if __name__ == '__main__':
     load_dotenv(env_path)
     os.environ['MONGO_HOST'] = 'localhost'
     # connect to database
-    collection, db, client = connect()
+    collection, db, client = connect_mongodb()
     print(client.server_info())
     # download images
     data = None

misc/model_outputs_from_annotation.py

@@ -0,0 +1,9 @@
+from shared.docstore.classes import ModelData
+
+if __name__ == '__main__':
+    # instantiate data object
+    vis_com_list = [ModelData.from_random() for i in range(3)]
+    # generate random predictions
+    [vis_com.from_random() for vis_com in vis_com_list]
+    for vis_com in vis_com_list:
+        print(vis_com)

misc/total_annotated.py

@@ -3,10 +3,10 @@ from __future__ import annotations
 import os
 from pathlib import Path
 
-from database import connect
-from database import total_annotated
 from dotenv import load_dotenv
 
+from shared.docstore import connect_mongodb, count_documents
+
 if __name__ == '__main__':
     # prepare env vars
     env_path = Path(__file__).parent.parent / 'local.env'
@@ -14,7 +14,10 @@ if __name__ == '__main__':
     load_dotenv(env_path)
     os.environ['MONGO_HOST'] = 'localhost'
     # connect to database
-    collection, db, client = connect()
+    collection, db, client = connect_mongodb()
     # get visual communication
-    num_docs = total_annotated(collection)
+    num_docs = count_documents(
+        collection=collection,
+        only_with_annotation=True,
+    )
     print(f"number of annotated  documents in database: {num_docs}")

misc/total_documents.py

@@ -3,10 +3,10 @@ from __future__ import annotations
 import os
 from pathlib import Path
 
-from database import connect
-from database import total_documents
 from dotenv import load_dotenv
 
+from shared.docstore import connect_mongodb, count_documents
+
 if __name__ == '__main__':
     # prepare env vars
     env_path = Path(__file__).parent.parent / 'local.env'
@@ -14,7 +14,10 @@ if __name__ == '__main__':
     load_dotenv(env_path)
     os.environ['MONGO_HOST'] = 'localhost'
     # connect to database
-    collection, db, client = connect()
+    collection, db, client = connect_mongodb()
     # get visual communication
-    num_docs = total_documents(collection)
+    num_docs = count_documents(
+        collection=collection,
+        only_with_annotation=False,
+    )
     print(f"total number of documents in database: {num_docs}")

model/docker-compose.server.yml

@@ -0,0 +1,16 @@
+services:
+  vcda_model:
+    image: vcda_model:test
+    container_name: vcda_model_alone
+    build:
+      context: ../
+      dockerfile: ./Dockerfile.model
+    env_file:
+      - ../server.env
+    environment:
+      - ENV=TEST
+    deploy:
+      resources:
+        limits:
+          cpus: '2.000'
+          memory: 8G

model/src/configs/resnet18.json

@@ -0,0 +1,14 @@
+{
+    "datasets": {
+        "train": {
+            "filelist": "datasets/train.csv"
+        },
+        "val": {
+            "filelist": "datasets/val.csv"
+        }
+    },
+    "backbone": {
+        "class": "model.src.models.VisualCommunicationModel",
+        "model_name": "pretrained"
+    }
+}

model/src/dataloader.py

@@ -1,151 +0,0 @@
-from __future__ import annotations
-
-import random
-from io import BytesIO
-
-from torch import Tensor
-from torch.utils.data import Dataset
-from torchvision.io import read_image
-from torchvision.transforms import ColorJitter
-from torchvision.transforms import InterpolationMode
-from torchvision.transforms import Normalize
-from torchvision.transforms.functional import hflip
-from torchvision.transforms.functional import pad
-from torchvision.transforms.functional import resize
-from torchvision.transforms.functional import rotate
-
-from core.database import connect
-from core.database import list_names
-
-# resnet18 original normalization values
-RESNET_NORMALIZE_MEAN = [0.485, 0.456, 0.406]
-RESNET_NORMALIZE_STD = [0.229, 0.224, 0.225]
-
-
-class VCDADataset(Dataset):
-    def __init__(
-        self,
-        do_augment: bool = False,
-        random_annotations: bool = False,
-        normalize_mean: list[float] = RESNET_NORMALIZE_MEAN,
-        normalize_std: list[float] = RESNET_NORMALIZE_STD,
-    ):
-        super().__init__()
-        self.do_augment = do_augment
-        self.random_annotations = random_annotations
-        self.normalize_mean = normalize_mean
-        self.normalize_std = normalize_std
-        # prepare augmentation functions
-        self.normalize = Normalize(
-            mean=normalize_mean,
-            std=normalize_std,
-        )
-        self.color_jitter = ColorJitter(
-            brightness=1e-1,
-            contrast=8e-2,
-            saturation=8e-2,
-        )
-        # connect to database
-        collection, _, _ = connect()
-        self.collection = collection
-        # load data names
-        has_annotation = False if self.random_annotations else True
-        self.data_name_list = list_names(
-            collection=self.collection,
-            has_annotation=has_annotation,
-        )
-
-    def __len__(self):
-        return len(self.data_name_list)
-
-    def __getitem__(self, idx):
-        # get image from database
-        name = self.data_name_list[idx]
-        query = {
-            'name': name,
-        }
-        projection = {
-            '_id': False,
-            'image': True,
-        }
-        img_bytes = self.collection.find_one(
-            filter=query,
-            projection=projection,
-        )
-        img = self.load_image(img_bytes)
-        if self.do_augment:
-            img = self.augment(img)
-        return img
-
-    def load_image(
-        self,
-        data: bytes,
-    ) -> Tensor:
-        """Load images tensor from bytes."""
-        assert isinstance(data, bytes)
-        img = read_image(BytesIO(data))
-        img /= 255  # normalize 8-bit image
-        img = self.square_pad(img)
-        img = resize(
-            img=img,
-            size=(512, 512),
-            interpolation=InterpolationMode.BICUBIC,
-        )
-        return img
-
-    @staticmethod
-    def square_pad(
-        img: Tensor,
-    ) -> Tensor:
-        """
-        Pads image to a square with side length
-        equal to the largest side of the input image.
-        """
-        assert isinstance(img, Tensor)
-        # B, nc, w, h = img.shape
-        h = img.shape[-2]
-        w = img.shape[-1]
-        if h == w:
-            return img
-        max_wh = max([h, w])
-        hp = int((max_wh - w) / 2)
-        vp = int((max_wh - h) / 2)
-        padding = (hp, vp, hp, vp)
-        return pad(img, padding, 0, 'constant')
-
-    def augment(
-        self,
-        img: Tensor,
-    ) -> Tensor:
-        """
-        Augment image with random horizontal flips,
-        rotations and color jitter.
-        """
-        assert isinstance(img, Tensor)
-        # left-right flip
-        if random.random() >= 0.5:
-            img = hflip(img)
-        # rotation
-        rnd = random.random()
-        if rnd < 0.25:
-            img = rotate(img, angle=90)
-        if rnd < 0.5:
-            img = rotate(img, angle=180)
-        if rnd < 0.75:
-            img = rotate(img, angle=270)
-        # color jitter
-        img = self.color_jitter(img)
-        return img
-
-    def reverse_normalise(
-        self,
-        img: Tensor,
-    ) -> Tensor:
-        """
-        Reverse normalization to get an image
-        that can be interpreted by humans.
-        """
-        assert isinstance(img, Tensor)
-        img *= Tensor(self.normalize_std).reshape((3, 1, 1))
-        img += Tensor(self.normalize_mean).reshape((3, 1, 1))
-        return img

model/src/loss_fn.py

@@ -1,7 +0,0 @@
-from __future__ import annotations
-
-import torch.nn as nn
-
-
-def setup_criterion():
-    return nn.MSELoss()

model/src/main.py

@@ -0,0 +1,43 @@
+"""Main script to be run by service."""
+
+import json
+import logging
+from traceback import print_exc
+
+import torch
+from models import VisualCommunicationModel
+from tqdm import tqdm
+from utils import DEVICE, VCDADataset, load_model
+
+from shared.utils import setup_logging
+
+if __name__ == '__main__':
+    # setup logging
+    setup_logging()
+    # instantiate model
+    model: VisualCommunicationModel = load_model()
+    model.eval()
+    # setup dataset
+    dataset = VCDADataset(
+        data_name_list=[
+            '02dbaf48d713e4e6d3a6b98fd2dc866e',
+        ],
+        do_augment=False,
+    )
+
+    # make prediction
+    with torch.no_grad():
+        for i in tqdm(range(len(dataset))):
+            try:
+                # get image
+                image = dataset[i]
+                image = torch.unsqueeze(image, 0)  # add artificial batch dimension
+                image = image.to(DEVICE)
+                # make prediction
+                pred: dict = model(image)
+            except Exception:
+                print_exc()
+                continue
+            else:
+                print(json.dumps(pred, indent=4))
+    logging.debug('finished')

model/src/model_name.txt

@@ -0,0 +1 @@
+2bbe2be17a663f4299657378ac5e993e

model/src/models/__init__.py

@@ -0,0 +1,3 @@
+from __future__ import annotations
+
+from .visual_communication import VisualCommunicationModel

model/src/models/angle.py

@@ -0,0 +1,6 @@
+from .fully_connected import FullyConnectedModel
+
+
+class AngleTail(FullyConnectedModel):
+    def __init__(self):
+        super().__init__(num_out_features=3)

model/src/models/contact.py

@@ -0,0 +1,6 @@
+from .fully_connected import FullyConnectedModel
+
+
+class ContactTail(FullyConnectedModel):
+    def __init__(self):
+        super().__init__(num_out_features=2)

model/src/models/distance.py

@@ -0,0 +1,6 @@
+from .fully_connected import FullyConnectedModel
+
+
+class DistanceTail(FullyConnectedModel):
+    def __init__(self):
+        super().__init__(num_out_features=3)

model/src/models/framing.py

@@ -0,0 +1,6 @@
+from .fully_connected import FullyConnectedModel
+
+
+class FramingTail(FullyConnectedModel):
+    def __init__(self):
+        super().__init__(num_out_features=4)

model/src/models/fully_connected.py

@@ -0,0 +1,24 @@
+from torch import nn
+
+
+class FullyConnectedModel(nn.Module):
+    """Fully connected layers model template for intrepreting feature space-
+    output from ResNet18 head."""
+
+    def __init__(self, num_out_features: int):
+        super().__init__()
+        # define layers
+        self.fc1 = nn.Linear(in_features=512, out_features=128)
+        self.af1 = nn.ReLU()
+        self.fc2 = nn.Linear(in_features=128, out_features=32)
+        self.af2 = nn.ReLU()
+        self.fc3 = nn.Linear(in_features=32, out_features=num_out_features)
+
+    def forward(self, x):
+        """Pass input through model."""
+        x = self.fc1(x)
+        x = self.af1(x)
+        x = self.fc2(x)
+        x = self.af2(x)
+        x = self.fc3(x)
+        return x

model/src/models/information_value.py

@@ -0,0 +1,6 @@
+from .fully_connected import FullyConnectedModel
+
+
+class InformationValueTail(FullyConnectedModel):
+    def __init__(self):
+        super().__init__(num_out_features=3)

model/src/models/modality_color.py

@@ -0,0 +1,6 @@
+from .fully_connected import FullyConnectedModel
+
+
+class ModalityColorTail(FullyConnectedModel):
+    def __init__(self):
+        super().__init__(num_out_features=3)

model/src/models/modality_depth.py

@@ -0,0 +1,6 @@
+from .fully_connected import FullyConnectedModel
+
+
+class ModalityDepthTail(FullyConnectedModel):
+    def __init__(self):
+        super().__init__(num_out_features=3)

model/src/models/modality_lighting.py

@@ -0,0 +1,6 @@
+from .fully_connected import FullyConnectedModel
+
+
+class ModalityLightingTail(FullyConnectedModel):
+    def __init__(self):
+        super().__init__(num_out_features=3)

model/src/models/point_of_view.py

@@ -0,0 +1,6 @@
+from .fully_connected import FullyConnectedModel
+
+
+class PointOfViewTail(FullyConnectedModel):
+    def __init__(self):
+        super().__init__(num_out_features=2)

model/src/models/resnet18_head.py

@@ -5,11 +5,15 @@ import torchvision
 
 
 class ResNet18Head(nn.Module):
-    def __init__(self):
+    def __init__(self, download_resnet_weights: bool = False):
         super().__init__()
         # copy out parts from ResNet18 with weights
+        if download_resnet_weights:
+            weights = torchvision.models.ResNet18_Weights.IMAGENET1K_V1
+        else:
+            weights = None
         resnet18 = torchvision.models.resnet18(
-            weights=torchvision.models.ResNet18_Weights.IMAGENET1K_V1,
+            weights=weights,
         )
         # save relevant layers
         self.conv1 = resnet18.conv1
@@ -21,7 +25,7 @@ class ResNet18Head(nn.Module):
         self.layer3 = resnet18.layer3
         self.layer4 = resnet18.layer4
         self.avgpool = resnet18.avgpool
-        self.flat = nn.Flatten()    # size 512
+        self.flat = nn.Flatten()  # size 512
 
     def forward(self, x):
         x = self.conv1(x)

model/src/models/salience.py

@@ -0,0 +1,6 @@
+from .fully_connected import FullyConnectedModel
+
+
+class SalienceTail(FullyConnectedModel):
+    def __init__(self):
+        super().__init__(num_out_features=5)

model/src/models/visual_communication.py

@@ -0,0 +1,58 @@
+"""Definition of VisualCommunicationModel class."""
+
+from __future__ import annotations
+
+from torch import nn
+
+from .angle import AngleTail
+from .contact import ContactTail
+from .distance import DistanceTail
+from .framing import FramingTail
+from .information_value import InformationValueTail
+from .modality_color import ModalityColorTail
+from .modality_depth import ModalityDepthTail
+from .modality_lighting import ModalityLightingTail
+from .point_of_view import PointOfViewTail
+from .resnet18_head import ResNet18Head
+from .salience import SalienceTail
+from .visual_syntax import VisualSyntaxTail
+
+
+class VisualCommunicationModel(nn.Module):
+    """Visual communication model."""
+
+    def __init__(self, download_resnet_weights: bool = False):
+        super().__init__()
+        # store other models
+        self.resnet_head = ResNet18Head(download_resnet_weights)
+        self.visual_syntax_tail = VisualSyntaxTail()
+        self.contact_tail = ContactTail()
+        self.angle_tail = AngleTail()
+        self.point_of_view_tail = PointOfViewTail()
+        self.distance_tail = DistanceTail()
+        self.modality_lighting_tail = ModalityLightingTail()
+        self.modality_color_tail = ModalityColorTail()
+        self.modality_depth_tail = ModalityDepthTail()
+        self.information_value_tail = InformationValueTail()
+        self.framing_tail = FramingTail()
+        self.salience_tail = SalienceTail()
+
+    def forward(self, x) -> dict:
+        """Calculate model output on data."""
+        # generate visual representation
+        features = self.resnet_head(x)
+        # make predictions
+        prediction_dict = {
+            'visual_syntax': self.visual_syntax_tail(features),
+            'contact': self.contact_tail(features),
+            'angle': self.angle_tail(features),
+            'point_of_view': self.point_of_view_tail(features),
+            'distance': self.distance_tail(features),
+            'modality_lighting': self.modality_lighting_tail(features),
+            'modality_color': self.modality_color_tail(features),
+            'modality_depth': self.modality_depth_tail(features),
+            'information_value': self.information_value_tail(features),
+            'framing': self.framing_tail(features),
+            'salience': self.salience_tail(features),
+        }
+        return prediction_dict

model/src/models/visual_syntax.py

@@ -0,0 +1,6 @@
+from .fully_connected import FullyConnectedModel
+
+
+class VisualSyntaxTail(FullyConnectedModel):
+    def __init__(self):
+        super().__init__(num_out_features=18)

model/src/train.py

@@ -1,6 +0,0 @@
-from __future__ import annotations
-
-from dataloader import VCDADataset  # noqa: F401
-from loss_fn import setup_criterion  # noqa: F401
-
-from model import ResNet18Head  # noqa: F401

model/src/utils/__init__.py

@@ -0,0 +1,3 @@
+from .get_class import get_class
+from .load_model import DEVICE, load_model
+from .vcda_dataset import VCDADataset

model/src/utils/get_class.py

@@ -0,0 +1,11 @@
+from importlib import import_module
+
+from torch.nn import Module
+
+
+def get_class(path: str) -> Module:
+    parts = path.split('.')
+    module_path = '.'.join(parts[:-1])
+    class_name = parts[-1]
+    module = import_module(module_path)
+    return getattr(module, class_name)

model/src/utils/get_model_name.py

@@ -0,0 +1,19 @@
+"""Definition of get_model_name function."""
+
+import logging
+from pathlib import Path
+
+
+def get_model_name(
+    path: Path,
+) -> str:
+    """Get model name from model_name file."""
+    assert isinstance(path, Path)
+    assert path.exists(), f'{path} does not exist'
+    # read file contents
+    with open(file=path, encoding='utf-8') as fh:
+        model_name = fh.read()
+    # remove newline character
+    model_name = model_name.strip()
+    logging.debug('finished')
+    return model_name

model/src/utils/load_model.py

@@ -0,0 +1,37 @@
+"""Definition of load_model function."""
+
+import logging
+from pathlib import Path
+
+import torch
+
+from model.src.models import VisualCommunicationModel
+from shared.repositories import ModelRepository
+
+from .get_model_name import get_model_name
+
+DEVICE = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+
+
+def load_model() -> VisualCommunicationModel:
+    """Instantiate model with weights loaded from latest model saved in
+    MinIO."""
+    # instantiate model
+    model = VisualCommunicationModel()
+    # get model object name
+    model_name_path = Path('model_name.txt')
+    model_object_name = get_model_name(path=model_name_path)
+    logging.info('using model: %s', model_object_name)
+    # load model data
+    with ModelRepository() as repo:
+        model_data = repo.get_data(model_object_name)
+    if model_data is None:
+        raise FileNotFoundError(f'model {model_object_name} not found')
+    model_checkpoint = torch.load(model_data.buffer)
+    model.load_state_dict(model_checkpoint)
+    # clean memory
+    model_checkpoint.clear()
+    # move model to selected device
+    model = model.to(DEVICE)
+    logging.debug('finished')
+    return model

model/src/utils/train.py

@@ -0,0 +1,49 @@
+from __future__ import annotations
+
+from datetime import datetime
+
+import torch
+import torch.nn as nn
+from dataloader import VCDADataset  # noqa: F401
+
+from ..models import VisualCommunicationModel
+
+PRE_WARMUP_LR = 1e-10
+POST_WARMUP_LR = 1e-5
+BATCH_SIZE = 32
+MODEL_NAME = 'visual_communication_model_v1'
+MAX_EPOCHS = 200
+RUN_NAME = datetime.now().strftime('%Y-%m-%d-%H%M') + '_' + MODEL_NAME
+DEVICE = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+
+
+# create loss function
+criterion = nn.MSELoss()
+
+
+def loss_fn(pred, target):
+    """Create loss function."""
+    return criterion(pred, target.unsqueeze(-1))
+
+
+# create model
+model = VisualCommunicationModel().to(DEVICE)
+
+# create optimizer
+optim = torch.optim.Adam(model.parameters(), lr=POST_WARMUP_LR)
+
+# create datasets
+# train_dataset
+# validation_dataset
+
+# create dataloaders
+
+# create trainer and evaluator
+
+# setup progressbar
+
+# setup lr scheduler
+
+# setup checkpoint saving
+
+# load in checkpoint if exists

model/src/utils/vcda_dataset.py

@@ -0,0 +1,134 @@
+"""Definition of Visual Critial Discourse Analysis dataloader."""
+
+import random
+
+from PIL import Image
+from torch import Tensor
+from torch.utils.data import Dataset
+from torchvision.transforms import ColorJitter, InterpolationMode, Normalize
+from torchvision.transforms.functional import (
+    hflip,
+    pad,
+    resize,
+    rotate,
+    to_pil_image,
+    to_tensor,
+)
+
+from shared.repositories import ImageRepository
+
+# resnet18 original normalization values
+RESNET_NORMALIZE_MEAN = [0.485, 0.456, 0.406]
+RESNET_NORMALIZE_STD = [0.229, 0.224, 0.225]
+
+
+class VCDADataset(Dataset):
+    """VCDA dataset class."""
+
+    def __init__(
+        self,
+        data_name_list: list[str],
+        do_augment: bool = False,
+        random_annotations: bool = False,
+    ):
+        super().__init__()
+        self.data_name_list = data_name_list
+        self.do_augment = do_augment
+        self.random_annotations = random_annotations
+        self.normalize_mean = RESNET_NORMALIZE_MEAN
+        self.normalize_std = RESNET_NORMALIZE_STD
+        # prepare augmentation functions
+        self.normalize = Normalize(
+            mean=self.normalize_mean,
+            std=self.normalize_std,
+        )
+        self.color_jitter = ColorJitter(
+            brightness=1e-1,
+            contrast=8e-2,
+            saturation=8e-2,
+        )
+
+    def __len__(self):
+        return len(self.data_name_list)
+
+    def __getitem__(self, idx):
+        # get image from database
+        image_name = self.data_name_list[idx]
+        with ImageRepository() as repo:
+            image_data = repo.get_data(image_name)
+        assert image_data is not None
+        tensor = self.image_to_tensor(image_data.image)
+        if self.do_augment:
+            tensor = self.augment(tensor)
+        return tensor
+
+    def image_to_tensor(
+        self,
+        image: Image.Image,
+    ) -> Tensor:
+        """Load images tensor from bytes."""
+        tensor = to_tensor(image)
+        tensor /= 255  # normalize 8-bit image
+        tensor = self.square_pad(tensor=tensor)
+        tensor = resize(
+            img=tensor,
+            size=(512, 512),
+            interpolation=InterpolationMode.BICUBIC,
+        )
+        return tensor
+
+    @staticmethod
+    def square_pad(
+        tensor: Tensor,
+    ) -> Tensor:
+        """Pads image to a square with side length equal to the largest side of
+        the input image."""
+        assert isinstance(tensor, Tensor)
+        # B, nc, w, h = img.shape
+        h = tensor.shape[-2]
+        w = tensor.shape[-1]
+        if h == w:
+            return tensor
+        max_wh = max([h, w])
+        hp = int((max_wh - w) / 2)
+        vp = int((max_wh - h) / 2)
+        padding = (hp, vp, hp, vp)
+        tensor = pad(tensor, padding, 0, 'constant')
+        return tensor
+
+    def augment(
+        self,
+        tensor: Tensor,
+    ) -> Tensor:
+        """Augment image with random horizontal flips, rotations and color
+        jitter."""
+        assert isinstance(tensor, Tensor)
+        # left-right flip
+        if random.random() >= 0.5:
+            tensor = hflip(tensor)
+        # rotation
+        rnd = random.random()
+        if rnd < 0.25:
+            tensor = rotate(tensor, angle=90)
+        if rnd < 0.5:
+            tensor = rotate(tensor, angle=180)
+        if rnd < 0.75:
+            tensor = rotate(tensor, angle=270)
+        # color jitter
+        tensor = self.color_jitter(tensor)
+        return tensor
+
+    def reverse_normalise(
+        self,
+        tensor: Tensor,
+    ) -> Image.Image:
+        """Reverse normalization to get an image that can be interpreted by
+        humans."""
+        assert isinstance(tensor, Tensor)
+        tensor *= Tensor(self.normalize_std).reshape((3, 1, 1))
+        tensor += Tensor(self.normalize_mean).reshape((3, 1, 1))
+        image = to_pil_image(
+            pic=tensor,
+            mode='RGB',
+        )
+        return image

model/train.py

@@ -0,0 +1,163 @@
+import json
+import os
+from argparse import ArgumentParser
+from datetime import datetime, timezone
+from pathlib import Path
+
+import torch
+from dotenv import load_dotenv
+from ignite.engine import Events, create_supervised_evaluator, create_supervised_trainer
+from ignite.handlers import global_step_from_engine
+from ignite.handlers.checkpoint import Checkpoint
+from ignite.handlers.param_scheduler import create_lr_scheduler_with_warmup
+from ignite.handlers.tensorboard_logger import TensorboardLogger
+from ignite.handlers.tqdm_logger import ProgressBar
+from ignite.metrics import Average, Loss, RunningAverage
+from torch import nn
+from torch.optim.lr_scheduler import ExponentialLR
+from torch.utils.data import DataLoader
+
+from model.src.utils import VCDADataset, get_class
+
+
+def parse_arguments():
+    parser = ArgumentParser()
+    parser.add_argument('-c', '--config', type=Path, required=True)
+    parser.add_argument('-d', '--device', type=str, default='cpu')
+    parser.add_argument('-b', '--batch-size', type=int, default=32)
+    parser.add_argument('--epochs', type=int, default=50)
+    parser.add_argument('--epoch-length', type=int, default=128)
+    parser.add_argument('--checkpoint', type=Path)
+    parser.add_argument('--lr', type=float, default=1e-5)
+    parser.add_argument('--lr-gamma', type=float, default=0.95)
+    parser.add_argument('--lr-warmup-start', type=float, default=1e-8)
+    parser.add_argument('--lr-warmup-duration', type=int, default=5)
+    parser.add_argument('--train-seed', type=int, default=1312)
+    parser.add_argument('--val-seed', type=int, default=1313)
+    parser.add_argument('--loader-workers', type=int, default=8)
+    return parser.parse_args()
+
+
+args = parse_arguments()
+
+load_dotenv('server.env')
+
+# read model config
+with open(args.config) as fh:
+    config = json.load(fh)
+model_name = args.config.stem
+run_name = datetime.now(timezone.utc).strftime('%Y-%m-%d-%H%M') + '_' + model_name
+device = torch.device(args.device)
+
+# create model
+model_class = get_class(config['backbone']['class'])
+model = model_class().to(device)
+optimizer = torch.optim.Adam(model.parameters(), lr=args.lr)
+loss_fn = nn.CrossEntropyLoss()
+
+# create datasets and loaders
+with open('model/src/dataset/train.csv', encoding='utf-8') as fh:
+    train_data_name_list = fh.read().split('\n')
+train_dataset = VCDADataset(
+    data_name_list=train_data_name_list,
+)
+train_loader = DataLoader(dataset=train_dataset, num_workers=args.loader_workers)
+
+with open('model/src/dataset/val.csv', encoding='utf-8') as fh:
+    val_data_name_list = fh.read().split('\n')
+val_dataset = VCDADataset(data_name_list=val_data_name_list)
+val_loader = DataLoader(dataset=val_dataset, num_workers=args.loader_workers)
+
+# create trainer and evaluator
+trainer = create_supervised_trainer(
+    model=model,
+    optimizer=optimizer,
+    loss_fn=loss_fn,
+    device=device,
+)
+Average(output_transform=lambda x: x).attach(trainer, name='loss')
+RunningAverage(output_transform=lambda x: x, alpha=0.5).attach(
+    trainer,
+    'running_avg_loss',
+)
+ProgressBar(desc='Train', ncols=80).attach(trainer, ['running_avg_loss'])
+
+val_metrics = {
+    'loss': Loss(loss_fn=loss_fn, device=device),
+}
+evaluator = create_supervised_evaluator(
+    model=model,
+    metrics=val_metrics,  # type: ignore
+    device=device,
+)
+ProgressBar(desc='Val', ncols=80).attach(evaluator)
+
+# create lr scheduler
+lr_scheduler = ExponentialLR(
+    optimizer=optimizer,
+    gamma=args.lr_gamma,
+)
+lr_handler = create_lr_scheduler_with_warmup(
+    lr_scheduler=lr_scheduler,
+    warmup_start_value=args.lr_warmup_start,
+    warmup_duration=args.lr_warmup_duration,
+    warmup_end_value=args.lr,
+)
+trainer.add_event_handler(Events.EPOCH_STARTED, lr_handler)
+
+
+# log metrics to TensorBoard
+@trainer.on(Events.EPOCH_COMPLETED)
+def evaluate():
+    evaluator.run(val_loader, epoch_length=args.val_epoch_length)
+
+
+tb_logger = TensorboardLogger(log_dir=f'runs/logs/{run_name}')
+tb_logger.attach_opt_params_handler(
+    engine=trainer,
+    event_name=Events.EPOCH_COMPLETED,
+    optimizer=optimizer,
+)
+for tag, engine in [('train', trainer), ('val', evaluator)]:
+    tb_logger.attach_output_handler(
+        engine,
+        event_name=Events.EPOCH_COMPLETED,
+        tag=tag,
+        metric_names='all',
+        global_step_transform=global_step_from_engine(trainer),
+    )
+
+# Set up checkpoint saving
+to_save = {
+    'model': model,
+    'optimizer': optimizer,
+    'trainer': trainer,
+    'lr_scheduler': lr_scheduler,
+    'lr_handler': lr_handler,
+}
+checkpoint_handler = Checkpoint(
+    to_save,
+    f'runs/checkpoints/{run_name}',
+    n_saved=3,
+    filename_prefix='best',
+    score_function=lambda engine: -engine.state.metrics['loss'],
+    score_name='neg_val_loss',
+    global_step_transform=global_step_from_engine(trainer),
+)
+evaluator.add_event_handler(Events.COMPLETED, checkpoint_handler)
+
+if args.checkpoint:
+    Checkpoint.load_objects(to_load=to_save, checkpoint=str(args.checkpoint))
+
+# save model config
+os.makedirs('runs/configs/', exist_ok=True)
+with open(f'runs/configs/{run_name}.json', 'w', encoding='utf-8') as fh:
+    json.dump(config, fh)
+
+# start training
+trainer.run(
+    train_loader,
+    max_epochs=args.epochs,
+    epoch_length=args.epoch_length,
+)
+tb_logger.close()

other/transfer_images_mongo_minio.py

@@ -0,0 +1,124 @@
+"""Script to move all images from MongoDB to MinIO."""
+
+from __future__ import annotations
+
+import logging
+from pathlib import Path
+
+from bson import ObjectId
+from dotenv import load_dotenv
+from pymongo.collection import Collection
+
+from shared.datastore import Datastore
+from shared.docstore import connect_mongodb
+from shared.docstore.src.classes import VisualCommunication
+from shared.utils import check_env, setup_logging
+from web_ui.src.main import NECESSARY_ENV_VAR_LIST
+
+
+def list_mongo_document_ids(
+    collection: Collection,
+) -> list[ObjectId]:
+    """Get list of all VisualCommunication documents in MongoDB."""
+    # prepare query
+    query = collection.find(
+        filter={},
+        projection={'_id': True},
+    )
+    # execute query
+    res_list = list(query)
+    logging.debug('got %s document(s)', len(res_list))
+    # convert result
+    id_list = [elem['_id'] for elem in res_list]
+    return id_list
+
+
+def get_visual_communication(
+    collection: Collection,
+    doc_id: ObjectId,
+) -> VisualCommunication:
+    """Get image from MongoDB."""
+    # prepare query
+    query = collection.find(
+        filter={'_id': doc_id},
+        projection={'_id': False},
+    )
+    # execute query
+    res_list = list(query)
+    logging.debug('got %s document(s)', len(res_list))
+    # convert result
+    vis_com = VisualCommunication.model_validate(res_list[0])
+    return vis_com
+
+
+def update_visual_communication(
+    collection: Collection,
+    vis_com: VisualCommunication,
+) -> None:
+    """Update VisualCommunication document in MongoDB."""
+    try:
+        collection.update_one(
+            filter={
+                'name': vis_com.name,
+            },
+            update={
+                '$set': vis_com.model_dump(),
+            },
+        )
+    except Exception as exc:
+        logging.error('failed saving %s', vis_com.name)
+        logging.debug(exc)
+        raise exc
+
+
+if __name__ == '__main__':
+    # load in env file
+    env_path = Path(__file__).parent.parent / 'server.env'
+    assert env_path.exists()
+    load_dotenv(env_path)
+    # ensure env vars set
+    check_env(NECESSARY_ENV_VAR_LIST)
+    # setup logging
+    setup_logging()
+    # connect to minIO
+    datastore = Datastore()
+    datastore.connect()
+    assert datastore._client is not None
+    # connect to MongoDB
+    collection, db, client = connect_mongodb()
+    # list documents in mongoDB
+    id_list = list_mongo_document_ids(collection)
+    for doc_id in id_list:
+        try:
+            # get image from MongoDB
+            vis_com = get_visual_communication(collection, doc_id)
+        except Exception as exc:
+            logging.debug(exc)
+            logging.error('failed getting image from document: %s', doc_id)
+            continue
+        try:
+            # get image
+            image = vis_com.get_image(
+                minio_client=datastore._client,
+            )
+            # put buffer in minio
+            object_name = datastore.put_image(
+                image=image,
+            )
+        except Exception as exc:
+            logging.debug(exc)
+            logging.error('failed saving image to minio')
+            continue
+        # update visual communication in mongodb
+        try:
+            vis_com.image = None  # type: ignore
+            vis_com.object_name = object_name
+            update_visual_communication(collection, vis_com)
+        except Exception as exc:
+            logging.debug(exc)
+            logging.error(
+                'failed updating visual communication %s',
+                vis_com.name,
+            )
+            continue
+        logging.debug('updated visual communication %s', vis_com.name)

pyproject.toml

@@ -5,26 +5,26 @@ description = ""
 authors = ["Brian Bjarke Jensen <bbj@skov.dk>"]
 readme = "README.md"
 packages = [
-    { include = "src" },
+    { include = "shared" },
 ]
 
 [tool.poetry.dependencies]
 python = "^3.12"
-gunicorn = "^21.2.0"
-python-dotenv = "^1.0.1"
-dash = "^2.15.0"
-dash-bootstrap-components = "^1.5.0"
-dash-mantine-components = "^0.12.1"
-pydantic = "^2.6.1"
-pillow = "^10.2.0"
-pymongo = "^4.6.1"
-dash-auth = "^2.2.0"
 
 
 [tool.poetry.group.test.dependencies]
 flake8 = "^7.0.0"
 mypy = "^1.8.0"
 types-pillow = "^10.2.0.20240213"
+types-requests = "^2.32.0.20240602"
+types-retry = "^0.9.9.4"
+flake8-pyproject = "^1.2.3"
+pandas-stubs = "^2.2.2.240603"
+types-tqdm = "^4.66.0.20240417"
+pytest = "^8.3.3"
+testcontainers = "^4.8.2"
+coverage = "^7.6.4"
+pytest-cov = "^6.0.0"
 
 
 [tool.poetry.group.dev.dependencies]
@@ -32,12 +32,93 @@ pandas = "^2.2.1"
 selenium = "^4.18.1"
 webdriver-manager = "^4.0.1"
 retry = "^0.9.2"
+pre-commit = "^4.1.0"
 
 
 [tool.poetry.group.model.dependencies]
-torch = "^2.2.1"
+torch = "^2.0.0"
 torchvision = "^0.17.1"
+torchinfo = "^1.8.0"
+minio = "^7.2.7"
+tqdm = "^4.66.4"
+pytorch-ignite = "^0.5.1"
+tensorboard = "^2.17.1"
+
+
+[tool.poetry.group.shared.dependencies]
+pymongo = "^4.7.2"
+pydantic = "^2.6.1"
+pillow = "^10.2.0"
+
+
+[tool.poetry.group.web_ui.dependencies]
+dash = "^2.17.0"
+gunicorn = "^21.2.0"
+dash-bootstrap-components = "^1.5.0"
+dash-mantine-components = "^0.12.1"
+dash-auth = "^2.2.0"
+
+
+[[tool.poetry.source]]
+name = "threadripper"
+url = "http://192.168.1.2:5001/index/"
+priority = "primary"
 
 [build-system]
 requires = ["poetry-core"]
 build-backend = "poetry.core.masonry.api"
+
+[tool.isort]
+profile = "black"
+
+[tool.flake8]
+per-file-ignores = "__init__.py:F401"
+max-line-length = 88
+extend-ignore = "E203"
+
+[tool.mypy]
+exclude = "image_download"
+
+[[tool.mypy.overrides]]
+module = "dash.*"
+ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = "dash_auth.*"
+ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = "dash_mantine_components.*"
+ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = "dash_bootstrap_components.*"
+ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = "torchvision.*"
+ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = "image_download.*"
+ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = "dataloader.*"
+ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = "utils.*"
+ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = "shared.datastore.*"
+ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = "models.*"
+ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = "shared.docstore.*"
+ignore_missing_imports = true

shared/repositories/__init__.py

@@ -0,0 +1,8 @@
+from .src import ImageRepository, ModelRepository, VisualCommunicationRepository
+from .src.dto import (
+    HexadecimalString,
+    ImageData,
+    ModelData,
+    VisualCommunicationData,
+    VisualCommunicationValues,
+)

shared/repositories/src/__init__.py

@@ -0,0 +1,3 @@
+from .image_repository import ImageRepository
+from .model_repository import ModelRepository
+from .visual_communication_repository import VisualCommunicationRepository

shared/repositories/src/dto/__init__.py

@@ -0,0 +1,7 @@
+from .hexadecimal_string import HexadecimalString
+from .image_data import ImageData
+from .model_data import ModelData
+from .visual_communication_data import (
+    VisualCommunicationData,
+    VisualCommunicationValues,
+)

shared/repositories/src/dto/bytes_io_pydantic_annotation.py

@@ -0,0 +1,37 @@
+"""Definition of BytesIO Pydantic Annotation."""
+
+from io import BytesIO
+from typing import Any
+
+from pydantic.json_schema import JsonSchemaValue
+from pydantic_core import core_schema
+
+
+class BytesIOPydanticAnnotation:
+    """Pydantic annotation that defines input validation, as well as general
+    and json serialization."""
+
+    @classmethod
+    def validate_input(cls, v: Any, handler) -> BytesIO:
+        """Pydantic-related function to validate input on instantiation."""
+        if isinstance(v, BytesIO):
+            return v
+        s = handler(v)
+        return BytesIO(s)
+
+    @classmethod
+    def __get_pydantic_core_schema__(
+        cls,
+        source_type,
+        _handler,
+    ) -> core_schema.CoreSchema:
+        assert source_type is BytesIO
+        return core_schema.no_info_wrap_validator_function(
+            function=cls.validate_input,
+            schema=core_schema.str_schema(),
+            serialization=core_schema.to_string_ser_schema(),
+        )
+
+    @classmethod
+    def __get_pydantic_json_schema__(cls, _core_schema, handler) -> JsonSchemaValue:
+        return handler(core_schema.str_schema())

shared/repositories/src/dto/hexadecimal_string.py

@@ -0,0 +1,24 @@
+"""Definition of Checksum DTO."""
+
+from __future__ import annotations
+
+import re
+
+
+class HexadecimalString(str):
+    """Hexadecimal-string class."""
+
+    def __new__(cls, string):
+        # ensure proper input format
+        pattern = r'[0-9-a-fA-F]{32}'
+        match = re.match(pattern, string)
+        if match is None:
+            raise ValueError(f'format does not match a hexadecimal-string: {string}')
+        return super().__new__(cls, string)
+
+    def __repr__(self) -> str:
+        class_name = self.__class__.__name__
+        return f"{class_name}('{self}')"
+
+    def __reduce__(self):
+        return self.__class__, (self,)

shared/repositories/src/dto/hexadecimal_string_pydantic_annotation.py

@@ -0,0 +1,38 @@
+"""Definition of HexadecimalString Pydantic Annotation."""
+
+from typing import Any
+
+from pydantic.json_schema import JsonSchemaValue
+from pydantic_core import core_schema
+
+from .hexadecimal_string import HexadecimalString
+
+
+class HexadecimalStringPydanticAnnotation:
+    """Pydantic annotation that defines input validation, as well as general
+    and json serialization."""
+
+    @classmethod
+    def validate_input(cls, v: Any, handler) -> HexadecimalString:
+        """Pydantic-related function to validate input on instantiation."""
+        if isinstance(v, HexadecimalString):
+            return v
+        s = handler(v)
+        return HexadecimalString(s)
+
+    @classmethod
+    def __get_pydantic_core_schema__(
+        cls,
+        source_type,
+        _handler,
+    ) -> core_schema.CoreSchema:
+        assert source_type is HexadecimalString
+        return core_schema.no_info_wrap_validator_function(
+            function=cls.validate_input,
+            schema=core_schema.str_schema(),
+            serialization=core_schema.to_string_ser_schema(),
+        )
+
+    @classmethod
+    def __get_pydantic_json_schema__(cls, _core_schema, handler) -> JsonSchemaValue:
+        return handler(core_schema.str_schema())

shared/repositories/src/dto/image_data.py

@@ -0,0 +1,16 @@
+"""Definition of VisualData DTO."""
+
+from typing import Annotated
+
+from PIL import Image
+from pydantic import Field
+
+from .image_pydantic_annotation import ImagePydanticAnnotation
+from .type_checking_base_model import TypeCheckingBaseModel
+
+
+class ImageData(TypeCheckingBaseModel):
+    """Visual data class."""
+
+    image: Annotated[Image.Image, ImagePydanticAnnotation]
+    name: str = Field(min_length=1)

shared/repositories/src/dto/image_pydantic_annotation.py

@@ -0,0 +1,37 @@
+"""Definition of BytesIO Pydantic Annotation."""
+
+from typing import Any
+
+from PIL import Image
+from pydantic.json_schema import JsonSchemaValue
+from pydantic_core import core_schema
+
+
+class ImagePydanticAnnotation:
+    """Pydantic annotation that defines input validation, as well as general
+    and json serialization."""
+
+    @classmethod
+    def validate_input(cls, v: Any, handler) -> Image.Image:
+        """Pydantic-related function to validate input on instantiation."""
+        if isinstance(v, Image.Image):
+            return v
+        s = handler(v)
+        return Image.open(s)
+
+    @classmethod
+    def __get_pydantic_core_schema__(
+        cls,
+        source_type,
+        _handler,
+    ) -> core_schema.CoreSchema:
+        assert source_type is Image.Image
+        return core_schema.no_info_wrap_validator_function(
+            function=cls.validate_input,
+            schema=core_schema.str_schema(),
+            serialization=core_schema.to_string_ser_schema(),
+        )
+
+    @classmethod
+    def __get_pydantic_json_schema__(cls, _core_schema, handler) -> JsonSchemaValue:
+        return handler(core_schema.str_schema())

shared/repositories/src/dto/model_data.py

@@ -0,0 +1,60 @@
+"""Definition of ModelData DTO."""
+
+from __future__ import annotations
+
+from hashlib import md5
+from io import BytesIO
+from typing import Annotated
+
+import torch
+from pydantic import Field
+
+from .bytes_io_pydantic_annotation import BytesIOPydanticAnnotation
+from .hexadecimal_string import HexadecimalString
+from .hexadecimal_string_pydantic_annotation import HexadecimalStringPydanticAnnotation
+from .type_checking_base_model import TypeCheckingBaseModel
+
+
+class ModelData(TypeCheckingBaseModel):
+    """Model Data DTO."""
+
+    buffer: Annotated[BytesIO, BytesIOPydanticAnnotation]
+    buffer_checksum: Annotated[HexadecimalString, HexadecimalStringPydanticAnnotation]
+    class_name: str = Field(
+        min_length=1,
+        description='name of model class to generate data.',
+    )
+
+    @staticmethod
+    def calculate_checksum(buffer: BytesIO) -> HexadecimalString:
+        """Calculate buffer checksum."""
+        checksum = md5(buffer.getbuffer()).hexdigest()
+        return HexadecimalString(checksum)
+
+    @staticmethod
+    def model_to_buffer(model: torch.nn.Module) -> BytesIO:
+        """Save model to buffer."""
+        assert isinstance(model, torch.nn.Module)
+        buffer = BytesIO()
+        torch.save(model.state_dict(), buffer)
+        return buffer
+
+    @classmethod
+    def from_model(cls, model: torch.nn.Module) -> ModelData:
+        """Instantiate from torch module."""
+        assert isinstance(model, torch.nn.Module)
+        # get model name
+        class_name = type(model).__name__
+        # save data to buffer
+        buffer = cls.model_to_buffer(model)
+        buffer = BytesIO()
+        torch.save(model.state_dict(), buffer)
+        # calculate checksum
+        buffer_checksum = cls.calculate_checksum(buffer)
+        # instantiate from buffer
+        data = cls(
+            buffer=buffer,
+            buffer_checksum=buffer_checksum,
+            class_name=class_name,
+        )
+        return data

shared/repositories/src/dto/type_checking_base_model.py

@@ -0,0 +1,11 @@
+"""Definition of TypeCheckingBaseModel class."""
+
+from pydantic import BaseModel, ConfigDict
+
+
+class TypeCheckingBaseModel(BaseModel):
+    """BaseModel with added type checking on input types."""
+
+    model_config = ConfigDict(
+        frozen=True,  # ensure data immutability
+    )

shared/repositories/src/dto/visual_communication_data/__init__.py

@@ -0,0 +1,2 @@
+from .visual_communication_data import VisualCommunicationData
+from .visual_communication_values import VisualCommunicationValues

shared/repositories/src/dto/visual_communication_data/angle_values.py

@@ -0,0 +1,11 @@
+"""Definition of AngleValues DTO."""
+
+from .values_model import ValuesModel
+
+
+class AngleValues(ValuesModel):
+    """Angle values DTO."""
+
+    high: float
+    eye_level: float
+    low: float

shared/repositories/src/dto/visual_communication_data/contact_values.py

@@ -0,0 +1,10 @@
+"""Definition of ContactValues DTO."""
+
+from .values_model import ValuesModel
+
+
+class ContactValues(ValuesModel):
+    """Contact values DTO."""
+
+    offer: float
+    demand: float

shared/repositories/src/dto/visual_communication_data/distance_values.py

@@ -0,0 +1,11 @@
+"""Definition of DistanceValues DTO."""
+
+from .values_model import ValuesModel
+
+
+class DistanceValues(ValuesModel):
+    """Distance values DTO."""
+
+    long: float
+    medium: float
+    close: float

shared/repositories/src/dto/visual_communication_data/framing_values.py

@@ -0,0 +1,12 @@
+"""Definition of FramingValues DTO."""
+
+from .values_model import ValuesModel
+
+
+class FramingValues(ValuesModel):
+    """Framing values DTO."""
+
+    frame_lines: float
+    empty_space: float
+    colour_contrast: float
+    form_contrast: float

shared/repositories/src/dto/visual_communication_data/information_value_values.py

@@ -0,0 +1,11 @@
+"""Definition of InformationValueValues DTO."""
+
+from .values_model import ValuesModel
+
+
+class InformationValueValues(ValuesModel):
+    """Information value values DTO."""
+
+    given_new: float
+    ideal_real: float
+    central_marginal: float

shared/repositories/src/dto/visual_communication_data/modality_color_values.py

@@ -0,0 +1,11 @@
+"""Definition of ModalityColorValues DTO."""
+
+from .values_model import ValuesModel
+
+
+class ModalityColorValues(ValuesModel):
+    """Modality color values DTO."""
+
+    high: float
+    medium: float
+    low: float

shared/repositories/src/dto/visual_communication_data/modality_depth_values.py

@@ -0,0 +1,11 @@
+"""Definition of ModalityDepthValues DTO."""
+
+from .values_model import ValuesModel
+
+
+class ModalityDepthValues(ValuesModel):
+    """Modality depth values DTO."""
+
+    high: float
+    medium: float
+    low: float

shared/repositories/src/dto/visual_communication_data/modality_lighting_values.py

@@ -0,0 +1,11 @@
+"""Definition of ModalityLightingValues DTO."""
+
+from .values_model import ValuesModel
+
+
+class ModalityLightingValues(ValuesModel):
+    """Modality lighting values DTO."""
+
+    high: float
+    medium: float
+    low: float

shared/repositories/src/dto/visual_communication_data/point_of_view_values.py

@@ -0,0 +1,10 @@
+"""Definition of PointOfViewValues DTO."""
+
+from .values_model import ValuesModel
+
+
+class PointOfViewValues(ValuesModel):
+    """Point-of-view values DTO."""
+
+    frontal: float
+    oblique: float

shared/repositories/src/dto/visual_communication_data/salience_values.py

@@ -0,0 +1,13 @@
+"""Definition of SalienceValues DTO."""
+
+from .values_model import ValuesModel
+
+
+class SalienceValues(ValuesModel):
+    """Salience values DTO."""
+
+    size: float
+    colour: float
+    tone: float
+    form: float
+    positioning: float

shared/repositories/src/dto/visual_communication_data/values_model.py

@@ -0,0 +1,60 @@
+"""Definition of ValuesModel base class."""
+
+from __future__ import annotations
+
+import random
+
+from pydantic import BaseModel, ConfigDict
+from torch import Tensor
+
+
+class ValuesModel(BaseModel):
+    """ValuesModel base class."""
+
+    model_config = ConfigDict(
+        validate_assignment=True,  # argument type checking
+        frozen=True,  # ensure data immutability
+    )
+
+    @classmethod
+    def list_fields(cls) -> list[str]:
+        """List options that are stored as attributes."""
+        return list(cls.model_fields.keys())
+
+    @classmethod
+    def from_random(cls):
+        """Instantiate with random numbers."""
+        kwargs = {field: random.random() for field in cls.list_fields()}
+        return cls(**kwargs)
+
+    @classmethod
+    def from_choice(cls, option: str) -> ValuesModel:
+        """Instantiate from choice."""
+        assert isinstance(option, str)
+        assert len(option) > 0
+        allowed_options_list = cls.list_fields()
+        if option not in allowed_options_list:
+            raise ValueError(f'option {option} must be in {allowed_options_list}')
+        # generate field values
+        kwargs = {field: 0 for field in allowed_options_list}
+        # set chosen value to max probability
+        kwargs[option] = 1
+        return cls(**kwargs)
+
+    @classmethod
+    def from_tensor(cls, tensor: Tensor):
+        """Instantiate from list of values."""
+        assert tensor.size(dim=0) == 1, f'tensor batch larger than 1: {tensor}'
+        data_list = [float(t.item()) for t in tensor[0]]
+        kwargs = dict(zip(cls.list_fields(), data_list))
+        return cls(**kwargs)
+
+    def highest_score_field(self) -> str:
+        """Return name of field with highest score."""
+        model_dict = self.model_dump()
+        return max(model_dict, key=lambda k: model_dict[k])
+
+    def highest_score_value(self) -> float:
+        """Return value of field with highest score."""
+        model_dict = self.model_dump()
+        return max(model_dict.values())

shared/repositories/src/dto/visual_communication_data/visual_communication_data.py

@@ -0,0 +1,14 @@
+"""Definition of VisualCommunicationData DTO."""
+
+from pydantic import Field
+
+from ..type_checking_base_model import TypeCheckingBaseModel
+from .visual_communication_values import VisualCommunicationValues
+
+
+class VisualCommunicationData(TypeCheckingBaseModel):
+    """Visual communication data class."""
+
+    name: str = Field(min_length=1)
+    annotation: VisualCommunicationValues | None = None
+    prediction: VisualCommunicationValues | None = None

shared/repositories/src/dto/visual_communication_data/visual_communication_values.py

@@ -0,0 +1,49 @@
+"""Definition of VisualCommunicationValues DTO."""
+
+from __future__ import annotations
+
+from ..type_checking_base_model import TypeCheckingBaseModel
+from .angle_values import AngleValues
+from .contact_values import ContactValues
+from .distance_values import DistanceValues
+from .framing_values import FramingValues
+from .information_value_values import InformationValueValues
+from .modality_color_values import ModalityColorValues
+from .modality_depth_values import ModalityDepthValues
+from .modality_lighting_values import ModalityLightingValues
+from .point_of_view_values import PointOfViewValues
+from .salience_values import SalienceValues
+from .visual_syntax_values import VisualSyntaxValues
+
+
+class VisualCommunicationValues(TypeCheckingBaseModel):
+    """Visual communication values class."""
+
+    visual_syntax: VisualSyntaxValues
+    contact: ContactValues
+    angle: AngleValues
+    point_of_view: PointOfViewValues
+    distance: DistanceValues
+    modality_lighting: ModalityLightingValues
+    modality_color: ModalityColorValues
+    modality_depth: ModalityDepthValues
+    information_value: InformationValueValues
+    framing: FramingValues
+    salience: SalienceValues
+
+    @classmethod
+    def from_random(cls) -> VisualCommunicationValues:
+        """Create a random instance."""
+        return cls(
+            visual_syntax=VisualSyntaxValues.from_random(),
+            contact=ContactValues.from_random(),
+            angle=AngleValues.from_random(),
+            point_of_view=PointOfViewValues.from_random(),
+            distance=DistanceValues.from_random(),
+            modality_lighting=ModalityLightingValues.from_random(),
+            modality_color=ModalityColorValues.from_random(),
+            modality_depth=ModalityDepthValues.from_random(),
+            information_value=InformationValueValues.from_random(),
+            framing=FramingValues.from_random(),
+            salience=SalienceValues.from_random(),
+        )

shared/repositories/src/dto/visual_communication_data/visual_syntax_values.py

@@ -0,0 +1,26 @@
+"""Definition of VisualSyntaxValues DTO."""
+
+from .values_model import ValuesModel
+
+
+class VisualSyntaxValues(ValuesModel):
+    """Visual syntax values DTO."""
+
+    non_transactional_action: float
+    non_transactional_reaction: float
+    unidirectional_transactional_action: float
+    unidirectional_transactional_reaction: float
+    bidirectional_transactional_action: float
+    bidirectional_transactional_reaction: float
+    conversion: float
+    speech_process: float
+    classification_overt_taxonomy: float
+    analytical_exhaustive: float
+    analytical_disarranged: float
+    analytical_temporal: float
+    analytical_distributed: float
+    analytical_topological: float
+    analytical_exploded: float
+    analytical_inclusive: float
+    symbolic_suggestive: float
+    symbolic_attributive: float

shared/repositories/src/image_repository.py

@@ -0,0 +1,74 @@
+"""Definition of ImageRepository class."""
+
+from __future__ import annotations
+
+from io import BytesIO
+from pathlib import Path
+
+from PIL import Image
+
+from .dto import ImageData
+from .implementations import MinioImplementation
+from .interfaces import ImageInterface
+
+
+class ImageRepository(ImageInterface, MinioImplementation):
+    """Image repository class that handles CRUD functionality for
+    VisualData."""
+
+    def __enter__(self) -> ImageRepository:
+        self.connect()
+        return self
+
+    @staticmethod
+    def _build_path(image_name: str) -> Path:
+        """Build object path."""
+        assert isinstance(image_name, str)
+        path = Path('images') / image_name
+        return path
+
+    def get_data(self, image_name: str) -> ImageData | None:
+        """Get Visual data."""
+        assert isinstance(image_name, str)
+        assert len(image_name) > 0
+        # build path
+        path = self._build_path(image_name)
+        # get object from bucket
+        buffer = self._get(path)
+        # handle if no data found
+        if not buffer:
+            return None
+        # convert data
+        image = Image.open(buffer)
+        data = ImageData(image=image, name=image_name)
+        return data
+
+    def put_data(self, data: ImageData) -> None:
+        """Put visual data."""
+        assert isinstance(data, ImageData)
+        # build path
+        path = self._build_path(data.name)
+        # save image to buffer
+        buffer = BytesIO()
+        data.image.save(buffer, 'png')
+        # put object in bucket
+        self._put(path, buffer)
+
+    def remove_data(self, image_name: str) -> None:
+        """Remove visual data."""
+        assert isinstance(image_name, str)
+        assert len(image_name) > 0
+        # build path
+        path = self._build_path(image_name)
+        # remove object
+        self._delete(path)
+
+    def list_names(self) -> list[str]:
+        """List names of all images."""
+        # build path
+        path = self._build_path('')
+        # list object paths
+        obj_path_list = self._list_objects(path)
+        # strip prefix
+        name_list = [obj_path.split('/')[-1] for obj_path in obj_path_list]
+        return name_list

shared/repositories/src/implementations/__init__.py

@@ -0,0 +1,2 @@
+from .minio_implementation import MinioImplementation
+from .mongo_implementation import MongoImplementation

shared/repositories/src/implementations/minio_implementation.py

@@ -0,0 +1,186 @@
+"""MinIO implementation."""
+
+from __future__ import annotations
+
+import logging
+import os
+from io import BytesIO
+from pathlib import Path
+from traceback import format_exc
+
+from minio import Minio
+
+from shared.utils import check_env
+
+from ..interfaces import DatabaseInterface
+
+
+class MinioImplementation(DatabaseInterface):
+    """MinIO basic CRUD implementation."""
+
+    def __init__(self):
+        # ensure necessary env vars available
+        var_list = {
+            'MINIO_ENDPOINT',
+            'MINIO_ACCESS_KEY',
+            'MINIO_SECRET_KEY',
+            'MINIO_BUCKET_NAME',
+        }
+        check_env(var_list)
+        # prepare internal variables
+        self._client: Minio | None = None
+        self._bucket_name: str | None = None
+
+    def connect(self):
+        """Connect to MinIO server."""
+        # prepare arguments
+        minio_endpoint = str(os.getenv('MINIO_ENDPOINT'))
+        minio_access_key = str(os.getenv('MINIO_ACCESS_KEY'))
+        minio_secret_key = str(os.getenv('MINIO_SECRET_KEY'))
+        minio_bucket_name = str(os.getenv('MINIO_BUCKET_NAME'))
+        # connect client
+        client = Minio(
+            endpoint=minio_endpoint,
+            access_key=minio_access_key,
+            secret_key=minio_secret_key,
+            secure=False,
+        )
+        # ensure bucket exists
+        if not client.bucket_exists(bucket_name=minio_bucket_name):
+            logging.debug('creating bucket: %s', minio_bucket_name)
+            client.make_bucket(bucket_name=minio_bucket_name)
+        # persist state
+        self._client = client
+        self._bucket_name = minio_bucket_name
+
+    def close(self) -> None:
+        """Close connection to MinIO server.
+
+        N.B. MinIO connection cannot be closed manually.
+        """
+        self._client = None
+        self._bucket_name = None
+
+    def connected(self):
+        """Check connection to Minio."""
+        res = isinstance(self._client, Minio)
+        logging.debug(res)
+        return res
+
+    def __enter__(self) -> MinioImplementation:
+        self.connect()
+        return self
+
+    def __exit__(self, exc_type, exc_val, exc_tb) -> None:
+        if any(
+            (
+                exc_type is not None,
+                exc_val is not None,
+                exc_tb is not None,
+            ),
+        ):
+            logging.error('error while exiting context')
+        self.close()
+
+    def _put(
+        self,
+        path: Path,
+        buffer: BytesIO,
+    ) -> None:
+        """Save in-memory buffer as object in MinIO."""
+        assert isinstance(path, Path)
+        assert isinstance(buffer, BytesIO)
+        assert isinstance(self._client, Minio)
+        assert isinstance(self._bucket_name, str)
+        # prepare for saving
+        num_bytes = len(buffer.getvalue())
+        buffer.seek(0)
+        # send data to bucket
+        try:
+            self._client.put_object(
+                bucket_name=self._bucket_name,
+                object_name=path.as_posix(),
+                length=num_bytes,
+                data=buffer,
+            )
+            logging.debug('saved data to %s', path)
+        except Exception as exc:
+            logging.error('failed saving data to MinIO')
+            raise exc
+
+    def _get(
+        self,
+        path: Path,
+    ) -> BytesIO | None:
+        """Get object from MinIO as in-memory buffer."""
+        assert isinstance(path, Path)
+        assert isinstance(self._client, Minio)
+        assert isinstance(self._bucket_name, str)
+        try:
+            # make request
+            response = self._client.get_object(
+                bucket_name=self._bucket_name,
+                object_name=path.as_posix(),
+            )
+            assert response.status == 200
+            # get buffer
+            buffer = BytesIO()
+            chunk_size = 2**14
+            while chunk := response.read(chunk_size):
+                buffer.write(chunk)
+            buffer.seek(0)
+            logging.debug('got %s', path)
+            return buffer
+        except Exception:
+            logging.error('failed getting data from MinIO')
+            logging.debug(format_exc())
+            return None
+        finally:
+            # close connection if established
+            if 'response' in locals():
+                response.close()
+                response.release_conn()
+
+    def _delete(
+        self,
+        path: Path,
+    ) -> None:
+        """Delete object from MinIO."""
+        assert isinstance(path, Path)
+        assert isinstance(self._client, Minio)
+        assert isinstance(self._bucket_name, str)
+        # remove object
+        try:
+            self._client.remove_object(
+                bucket_name=self._bucket_name,
+                object_name=path.as_posix(),
+            )
+            logging.debug('deleted %s', path)
+        except Exception as exc:
+            logging.error('failed deleting %s', path)
+            logging.debug(format_exc())
+            raise exc
+
+    def _list_objects(
+        self,
+        path: Path,
+    ) -> list[str]:
+        """List objects in bucket under path."""
+        assert isinstance(path, Path)
+        assert isinstance(self._client, Minio)
+        assert isinstance(self._bucket_name, str)
+        try:
+            # list objects
+            obj_list = self._client.list_objects(
+                bucket_name=self._bucket_name,
+                prefix=path.as_posix(),
+                recursive=True,
+            )
+            # extract info
+            name_list = [obj.object_name for obj in obj_list]
+            logging.debug('got %s objects matching %s', len(name_list), path)
+            return name_list
+        except Exception as exc:
+            logging.error('failed listing objects under %s', path)
+            logging.debug(format_exc())
+            raise exc

shared/repositories/src/implementations/mongo_implementation.py

@@ -0,0 +1,145 @@
+"""Mongo implementation."""
+
+from __future__ import annotations
+
+import logging
+import os
+import traceback
+
+from pymongo import MongoClient
+from pymongo.collection import Collection
+from pymongo.database import Database
+from pymongo.errors import ServerSelectionTimeoutError
+
+from shared.utils import check_env
+
+from ..interfaces import DatabaseInterface
+
+
+class MongoImplementation(DatabaseInterface):
+    """MongoDB basic CRUD implementation."""
+
+    def __init__(self):
+        # ensure necessary env vars available
+        var_list = {
+            'MONGO_ENDPOINT',
+            'MONGO_DB',
+            'MONGO_COLLECTION',
+        }
+        check_env(var_list)
+        # prepare internal variables
+        self.client: MongoClient | None = None
+        self.db: Database | None = None
+        self.collection: Collection | None = None
+
+    def connect(self) -> None:
+        """Connect to Mongo server."""
+        # prepare arguments
+        mongo_endpoint = str(os.getenv('MONGO_ENDPOINT'))
+        mongo_database = str(os.getenv('MONGO_DB'))
+        mongo_collection = str(os.getenv('MONGO_COLLECTION'))
+        # connect client
+        client: MongoClient = MongoClient(mongo_endpoint)
+        database = client[mongo_database]
+        collection = database[mongo_collection]
+        # set unique index on 'name'
+        collection.create_index(keys='name', unique=True)
+        # persist state
+        self._client = client
+        self._database = database
+        self._collection = collection
+
+    def close(self) -> None:
+        """Close connection to Mongo server."""
+        self._client.close()
+        self._client = None  # type: ignore
+        self._database = None  # type: ignore
+        self._collection = None  # type: ignore
+
+    def connected(self) -> bool:
+        """Check connection to Mongo."""
+        if self._client is None:
+            return False
+        try:
+            # trigger fetch data
+            _ = self._client.server_info()
+            res = True
+        except ServerSelectionTimeoutError:
+            res = False
+        logging.debug(res)
+        return res
+
+    def __enter__(self) -> MongoImplementation:
+        self.connect()
+        return self
+
+    def __exit__(self, exc_type, exc_val, exc_tb) -> None:
+        if any(
+            (
+                exc_type is not None,
+                exc_val is not None,
+                exc_tb is not None,
+            ),
+        ):
+            logging.error('error while exiting context')
+            traceback.print_exception(exc_type, exc_val, exc_tb)
+        self.close()
+
+    def _save(
+        self,
+        data: dict,
+        query: dict,
+    ) -> None:
+        """Save document in Mongo."""
+        assert isinstance(data, dict)
+        assert isinstance(query, dict)
+        assert 'name' in data
+        assert self.connected()
+        self._collection.update_one(
+            filter=query,
+            update={
+                '$set': data.copy(),
+            },
+            upsert=True,
+        )
+        logging.debug('Save %s', data)
+
+    def _get(
+        self,
+        query: dict,
+    ) -> dict | None:
+        """Get document from Mongo."""
+        assert isinstance(query, dict)
+        assert self.connected()
+        doc = self._collection.find_one(query, projection={'_id': False})
+        logging.debug('Found %s', doc)
+        return doc
+
+    def _delete(
+        self,
+        query: dict,
+    ) -> None:
+        """Remove document from Mongo."""
+        assert isinstance(query, dict)
+        assert self.connected()
+        doc = self._collection.delete_one(query)
+        logging.debug('Deleted %s', doc)
+
+    def _list_documents(self, key='name') -> list[str]:
+        """List documents in Mongo."""
+        assert isinstance(key, str)
+        assert len(key) > 0
+        # build query
+        doc_list = list(
+            self._collection.find(
+                filter={},
+                projection={
+                    '_id': False,
+                    key: True,
+                },
+            ),
+        )
+        # extract values
+        value_list = [doc[key] for doc in doc_list]
+        logging.debug('Got %s document(s)', len(doc_list))
+        return value_list

shared/repositories/src/interfaces/__init__.py

@@ -0,0 +1,4 @@
+from .database_interface import DatabaseInterface
+from .image_interface import ImageInterface
+from .model_interface import ModelInterface
+from .visual_communication_interface import VisualCommunicationInterface

shared/repositories/src/interfaces/database_interface.py

@@ -0,0 +1,28 @@
+"""Definition of DatabaseInterface class."""
+
+from abc import ABC, abstractmethod
+
+
+class DatabaseInterface(ABC):
+    """Interface base class adding 'connect', 'close' and context
+    functionalities."""
+
+    @abstractmethod
+    def connect(self):
+        raise NotImplementedError()
+
+    @abstractmethod
+    def close(self):
+        raise NotImplementedError()
+
+    @abstractmethod
+    def connected(self) -> bool:
+        raise NotImplementedError()
+
+    @abstractmethod
+    def __enter__(self):
+        raise NotImplementedError()
+
+    @abstractmethod
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        raise NotImplementedError()

shared/repositories/src/interfaces/image_interface.py

@@ -0,0 +1,26 @@
+"""Definition of ImageInterface."""
+
+from abc import abstractmethod
+
+from ..dto import ImageData
+from .database_interface import DatabaseInterface
+
+
+class ImageInterface(DatabaseInterface):
+    """Image interface class."""
+
+    @abstractmethod
+    def get_data(self, image_name: str) -> ImageData | None:
+        raise NotImplementedError()
+
+    @abstractmethod
+    def put_data(self, data: ImageData) -> None:
+        raise NotImplementedError()
+
+    @abstractmethod
+    def remove_data(self, image_name: str) -> None:
+        raise NotImplementedError()
+
+    @abstractmethod
+    def list_names(self) -> list[str]:
+        raise NotImplementedError()

shared/repositories/src/interfaces/model_interface.py

@@ -0,0 +1,26 @@
+"""Definition of ModelInterface."""
+
+from abc import abstractmethod
+
+from ..dto import ModelData
+from .database_interface import DatabaseInterface
+
+
+class ModelInterface(DatabaseInterface):
+    """Model interface class."""
+
+    @abstractmethod
+    def get_data(self, object_name: str) -> ModelData | None:
+        raise NotImplementedError()
+
+    @abstractmethod
+    def put_data(self, data: ModelData) -> None:
+        raise NotImplementedError()
+
+    @abstractmethod
+    def remove_data(self, object_name: str) -> None:
+        raise NotImplementedError()
+
+    @abstractmethod
+    def list_names(self) -> list[str]:
+        raise NotImplementedError()

shared/repositories/src/interfaces/visual_communication_interface.py

@@ -0,0 +1,26 @@
+"""Definition of VisualCommunicationInterface."""
+
+from abc import abstractmethod
+
+from ..dto import VisualCommunicationData
+from .database_interface import DatabaseInterface
+
+
+class VisualCommunicationInterface(DatabaseInterface):
+    """Visual communication interface class."""
+
+    @abstractmethod
+    def get_data(self, name: str) -> VisualCommunicationData | None:
+        raise NotImplementedError()
+
+    @abstractmethod
+    def put_data(self, data: VisualCommunicationData) -> None:
+        raise NotImplementedError()
+
+    @abstractmethod
+    def remove_data(self, name: str) -> None:
+        raise NotImplementedError()
+
+    @abstractmethod
+    def list_names(self) -> list[str]:
+        raise NotImplementedError()

shared/repositories/src/model_repository.py

@@ -0,0 +1,77 @@
+"""Definition of ModelRepository class."""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+from .dto import HexadecimalString, ModelData
+from .implementations import MinioImplementation
+from .interfaces import ModelInterface
+
+
+class ModelRepository(ModelInterface, MinioImplementation):
+    """Model repository class that handles CRUD functionality for ModelData."""
+
+    def __enter__(self) -> ModelRepository:
+        self.connect()
+        return self
+
+    @staticmethod
+    def _prefix() -> Path:
+        """Object name prefix."""
+        return Path('models')
+
+    @classmethod
+    def _build_object_name(cls, data: ModelData) -> str:
+        """Build object name from data."""
+        return f'{data.class_name}-{data.buffer_checksum}'
+
+    def get_data(self, object_name: str) -> ModelData | None:
+        """Get model data."""
+        assert isinstance(object_name, str)
+        # build path
+        path = self._prefix() / object_name
+        # get object from bucket
+        buffer = self._get(path)
+        # handle if no data found
+        if not buffer:
+            return None
+        # extract info
+        class_name, buffer_checksum_str = object_name.split('-')
+        # convert data
+        buffer_checksum = HexadecimalString(buffer_checksum_str)
+        # instantiate data
+        data = ModelData(
+            buffer=buffer,
+            buffer_checksum=buffer_checksum,
+            class_name=class_name,
+        )
+        return data
+
+    def put_data(self, data: ModelData) -> None:
+        """Put model data."""
+        assert isinstance(data, ModelData)
+        # build object name
+        object_name = self._build_object_name(data)
+        # build path
+        path = self._prefix() / object_name
+        # put object in bucket
+        self._put(path, data.buffer)
+
+    def remove_data(self, object_name: str) -> None:
+        """Remove model data."""
+        assert isinstance(object_name, str)
+        # build path
+        path = self._prefix() / object_name
+        # remove object
+        self._delete(path)
+
+    def list_names(self) -> list[str]:
+        """List names of all models."""
+        # build path
+        path = self._prefix()
+        # list object paths
+        obj_path_list = self._list_objects(path)
+        # strip prefix
+        name_list = [obj_path.split('/')[-1] for obj_path in obj_path_list]
+        return name_list

shared/repositories/src/visual_communication_repository.py

@@ -0,0 +1,56 @@
+"""Definition of VisualCommunicationRepository class."""
+
+from __future__ import annotations
+
+from .dto import VisualCommunicationData
+from .implementations import MongoImplementation
+from .interfaces import VisualCommunicationInterface
+
+
+class VisualCommunicationRepository(VisualCommunicationInterface, MongoImplementation):
+    """Visual communication repository class that handles CRUD functionality
+    for VisualCommunicationData."""
+
+    def __enter__(self) -> VisualCommunicationRepository:
+        self.connect()
+        return self
+
+    def get_data(self, name: str) -> VisualCommunicationData | None:
+        """Get visual communication data."""
+        assert isinstance(name, str)
+        assert len(name) > 0
+        # build query
+        query = {'name': name}
+        # get document from mongo
+        doc = self._get(query)
+        # handle if no data found
+        if doc is None:
+            return None
+        # instantiate object
+        data = VisualCommunicationData(**doc)
+        return data
+
+    def put_data(self, data: VisualCommunicationData) -> None:
+        """Put visual communication data."""
+        assert isinstance(data, VisualCommunicationData)
+        # convert to dict
+        data_dict: dict = data.model_dump(mode='json')
+        # build query
+        query = {'name': data.name}
+        # save to mongo
+        self._save(data_dict, query)
+
+    def remove_data(self, name: str) -> None:
+        """Remove visual communication data."""
+        assert isinstance(name, str)
+        assert len(name) > 0
+        # build query
+        query = {'name': name}
+        # remove document from mongo
+        self._delete(query)
+
+    def list_names(self) -> list[str]:
+        """List names of all documents."""
+        # list documents
+        name_list = self._list_documents(key='name')
+        return name_list

shared/repositories/tests/integration/conftest.py

@@ -0,0 +1,403 @@
+"""Integration tests configuration."""
+
+import os
+import random
+from collections.abc import Iterator
+from io import BytesIO
+from pathlib import Path
+
+import minio
+import pytest
+from dotenv import load_dotenv
+from PIL import Image
+from pymongo import MongoClient
+
+from model.src.models import VisualCommunicationModel
+from shared.repositories import (
+    HexadecimalString,
+    ImageData,
+    ImageRepository,
+    ModelData,
+    ModelRepository,
+    VisualCommunicationData,
+    VisualCommunicationRepository,
+    VisualCommunicationValues,
+)
+from shared.repositories.src.implementations import (
+    MinioImplementation,
+    MongoImplementation,
+)
+
+# set random seed for reproducibility
+random.seed(13)
+
+# define test environment variables
+necessary_env_vars = {
+    'MINIO_ENDPOINT',
+    'MINIO_ACCESS_KEY',
+    'MINIO_SECRET_KEY',
+    'MONGO_ENDPOINT',
+}
+env_var_map = {
+    'MINIO_BUCKET_NAME': 'test-bucket',
+    'MINIO_OBJECT_NAME': 'test-object',
+    'MINIO_IMAGE_NAME': 'test-image',
+    'MONGO_DB': 'test-db',
+    'MONGO_COLLECTION': 'test-collection',
+}
+
+
+@pytest.fixture(scope='session', autouse=True)
+def setup_env(
+    request: pytest.FixtureRequest,
+) -> None:
+    """Populate environment with variables used for testing."""
+    # load in optional local test environment variables
+    test_env_path = Path(__file__).parent.parent.parent.parent.parent / 'test.env'
+    load_dotenv(test_env_path)
+    # check if necessary env vars are set
+    for key in necessary_env_vars:
+        assert key in os.environ, f'{key} not set'
+    # set env vars unique to this test
+    for key, val in env_var_map.items():
+        # set env var
+        os.environ[key] = val
+
+    # ensure cleanup
+    def cleanup_env():
+        for key in env_var_map:
+            _ = os.environ.pop(key, default=None)
+
+    request.addfinalizer(cleanup_env)
+
+
+@pytest.fixture(scope='session')
+def raw_minio_client(
+    setup_env,
+) -> Iterator[minio.Minio]:
+    """Raw Minio client fixture."""
+    # prepare arguments
+    minio_endpoint = str(os.getenv('MINIO_ENDPOINT'))
+    minio_access_key = str(os.getenv('MINIO_ACCESS_KEY'))
+    minio_secret_key = str(os.getenv('MINIO_SECRET_KEY'))
+    minio_bucket_name = str(os.getenv('MINIO_BUCKET_NAME'))
+    # connect client
+    client = minio.Minio(
+        endpoint=minio_endpoint,
+        access_key=minio_access_key,
+        secret_key=minio_secret_key,
+        secure=False,
+    )
+    # ensure bucket exists
+    if not client.bucket_exists(bucket_name=minio_bucket_name):
+        client.make_bucket(bucket_name=minio_bucket_name)
+    # expose client
+    yield client
+    # cleanup
+    object_list = client.list_objects(minio_bucket_name, recursive=True)
+    for obj in object_list:
+        client.remove_object(
+            bucket_name=obj.bucket_name,
+            object_name=obj.object_name,
+        )
+    client.remove_bucket(minio_bucket_name)
+
+
+@pytest.fixture
+def minio_client(
+    setup_env,
+) -> Iterator[MinioImplementation]:
+    """MinioImplementation fixture."""
+    # instantiate and connect client
+    minio_client = MinioImplementation()
+    minio_client.connect()
+    # expose client
+    yield minio_client
+    # cleanup
+    object_name_list = minio_client._list_objects(Path('*'))
+    for name in object_name_list:
+        minio_client._delete(Path(name))
+    minio_client.close()
+
+
+@pytest.fixture
+def buffer() -> Iterator[BytesIO]:
+    """Bytes buffer fixture."""
+    # generate reproducible random data
+    data = random.randbytes(n=2**21)  # 2 MB
+    # convert data
+    buffer = BytesIO(data)
+    # expose buffer
+    yield buffer
+
+
+@pytest.fixture
+def buffer_in_minio(
+    raw_minio_client: minio.Minio,
+    buffer: BytesIO,
+) -> Iterator[tuple[Path, BytesIO]]:
+    """Buffer in Minio fixture."""
+    # prepare arguments
+    minio_bucket_name = str(os.getenv('MINIO_BUCKET_NAME'))
+    minio_object_name = str(os.getenv('MINIO_OBJECT_NAME'))
+    # prepare for saving
+    num_bytes = len(buffer.getvalue())
+    buffer.seek(0)
+    # put data in bucket
+    raw_minio_client.put_object(
+        bucket_name=minio_bucket_name,
+        object_name=minio_object_name,
+        length=num_bytes,
+        data=buffer,
+    )
+    # expose data
+    yield Path(minio_object_name), buffer
+    # cleanup
+    raw_minio_client.remove_object(
+        bucket_name=minio_bucket_name,
+        object_name=minio_object_name,
+    )
+
+
+@pytest.fixture
+def image_data() -> Iterator[ImageData]:
+    """Image data fixture."""
+    # prepare arguments
+    name = str(os.getenv('MINIO_IMAGE_NAME'))
+    image = Image.new(mode='RGB', size=(480, 480))
+    # instantiate data
+    image_data = ImageData(image=image, name=name)
+    # expose data
+    yield image_data
+
+
+@pytest.fixture
+def image_data_in_minio(
+    raw_minio_client: minio.Minio,
+    image_data: ImageData,
+) -> Iterator[ImageData]:
+    """Image data in Minio fixture."""
+    # prepare arguments
+    minio_bucket_name = str(os.getenv('MINIO_BUCKET_NAME'))
+    image_name = image_data.name
+    # build object path
+    object_path = ImageRepository._build_path(image_name)
+    # save image to buffer
+    buffer = BytesIO()
+    image_data.image.save(buffer, 'png')
+    # prepare for saving
+    num_bytes = len(buffer.getvalue())
+    buffer.seek(0)
+    # put data in bucket
+    raw_minio_client.put_object(
+        bucket_name=minio_bucket_name,
+        object_name=object_path.as_posix(),
+        length=num_bytes,
+        data=buffer,
+    )
+    # expose data
+    yield image_data
+    # cleanup
+    raw_minio_client.remove_object(
+        bucket_name=minio_bucket_name,
+        object_name=object_path.as_posix(),
+    )
+
+
+@pytest.fixture(scope='session')
+def image_repo(
+    setup_env,
+) -> Iterator[ImageRepository]:
+    """Image repository fixture."""
+    # define test environment variables
+    assert 'MINIO_ENDPOINT' in os.environ, 'MINIO_ENDPOINT not set'
+    assert 'MINIO_ACCESS_KEY' in os.environ, 'MINIO_ACCESS_KEY not set'
+    assert 'MINIO_SECRET_KEY' in os.environ, 'MINIO_SECRET_KEY not set'
+    assert 'MONGO_ENDPOINT' in os.environ, 'MONGO_ENDPOINT not set'
+    repo = ImageRepository()
+    repo.connect()
+    yield repo
+    repo.close()
+
+
+@pytest.fixture
+def model_data() -> Iterator[ModelData]:
+    """Model data fixture."""
+    # prepare arguments
+    vis_com_model = VisualCommunicationModel().to('cpu')
+    class_name = type(vis_com_model).__name__
+    buffer = ModelData.model_to_buffer(vis_com_model)
+    buffer_checksum = HexadecimalString('77dcab1769563654a6e24f92d40f29bd')
+    # instantiate data
+    model_data = ModelData(
+        buffer=buffer,
+        buffer_checksum=buffer_checksum,
+        class_name=class_name,
+    )
+    # expose model
+    yield model_data
+
+
+@pytest.fixture
+def model_data_in_minio(
+    raw_minio_client: minio.Minio,
+    model_data: ModelData,
+) -> Iterator[ModelData]:
+    """Model data in Minio fixture."""
+    # prepare arguments
+    minio_bucket_name = str(os.getenv('MINIO_BUCKET_NAME'))
+    object_name = ModelRepository._build_object_name(model_data)
+    buffer = model_data.buffer
+    # build object path
+    object_path = ModelRepository._prefix() / object_name
+    # prepare for saving
+    num_bytes = len(buffer.getvalue())
+    buffer.seek(0)
+    # put data in bucket
+    raw_minio_client.put_object(
+        bucket_name=minio_bucket_name,
+        object_name=object_path.as_posix(),
+        length=num_bytes,
+        data=buffer,
+    )
+    # expose data
+    yield model_data
+    # cleanup
+    raw_minio_client.remove_object(
+        bucket_name=minio_bucket_name,
+        object_name=object_path.as_posix(),
+    )
+
+
+@pytest.fixture(scope='session')
+def model_repo(
+    setup_env,
+) -> Iterator[ModelRepository]:
+    """Model repository fixture."""
+    repo = ModelRepository()
+    repo.connect()
+    yield repo
+    repo.close()
+
+
+@pytest.fixture
+def raw_mongo_client(
+    setup_env,
+) -> Iterator[MongoClient]:
+    """Raw mongo client fixture."""
+    # prepare arguments
+    mongo_endpoint = str(os.getenv('MONGO_ENDPOINT'))
+    mongo_database = str(os.getenv('MONGO_DB'))
+    # connect client
+    client: MongoClient = MongoClient(mongo_endpoint)
+    _ = client[mongo_database]
+    # expose client
+    yield client
+    # cleanup
+    client.drop_database(mongo_database)
+
+
+@pytest.fixture
+def mongo_client(
+    setup_env,
+) -> Iterator[MongoImplementation]:
+    """MongoImplementation fixture."""
+    # instantiate and connect client
+    mongo_client = MongoImplementation()
+    mongo_client.connect()
+    # expose client
+    yield mongo_client
+    # cleanup
+    mongo_client._collection.drop()
+    mongo_client.close()
+
+
+@pytest.fixture
+def dictionary() -> Iterator[dict]:
+    """Dictionary fixture."""
+    # prepare data
+    data = {
+        'name': 'test-dictionary',
+        'str_key': 'value',
+        'int_key': 100,
+        'float_key': 3.14,
+        'list_key': [1, 2, 3],
+        'dict_key': {
+            'nested_key': 'nested_value',
+        },
+    }
+    # expose data
+    yield data
+
+
+@pytest.fixture
+def dictionary_in_mongo(
+    raw_mongo_client: MongoClient,
+    dictionary: dict,
+) -> Iterator[dict]:
+    """Dictionary in Mongo fixture."""
+    # prepare arguments
+    database = str(os.getenv('MONGO_DB'))
+    collection = str(os.getenv('MONGO_COLLECTION'))
+    # save data
+    _ = raw_mongo_client[database][collection].insert_one(dictionary.copy())
+    # expose data
+    yield dictionary
+    # cleanup
+    raw_mongo_client[database][collection].delete_one(dictionary)
+
+
+@pytest.fixture
+def visual_communication_values() -> Iterator[VisualCommunicationValues]:
+    """Visual communication values fixture."""
+    # instantiate with random values
+    visual_communication_values = VisualCommunicationValues.from_random()
+    # expose values
+    yield visual_communication_values
+
+
+@pytest.fixture
+def visual_communication_data(
+    visual_communication_values: VisualCommunicationValues,
+) -> Iterator[VisualCommunicationData]:
+    """Visual communication data fixture."""
+    # prepare arguments
+    name = 'test-visual-communication'
+    annotation = visual_communication_values
+    # instantiate data
+    visual_communication_data = VisualCommunicationData(
+        name=name,
+        annotation=annotation,
+    )
+    # expose data
+    yield visual_communication_data
+
+
+@pytest.fixture
+def visual_communication_data_in_mongo(
+    raw_mongo_client: MongoClient,
+    visual_communication_data: VisualCommunicationData,
+) -> Iterator[VisualCommunicationData]:
+    """Visual communication data in Mongo fixture."""
+    # prepare arguments
+    database = str(os.getenv('MONGO_DB'))
+    collection = str(os.getenv('MONGO_COLLECTION'))
+    # convert data
+    dictionary = visual_communication_data.model_dump(mode='dict')
+    # save data
+    _ = raw_mongo_client[database][collection].insert_one(dictionary.copy())
+    # expose data
+    yield visual_communication_data
+    # cleanup
+    raw_mongo_client[database][collection].delete_one(dictionary)
+
+
+@pytest.fixture(scope='session')
+def visual_communication_repo(
+    setup_env,
+) -> Iterator[VisualCommunicationRepository]:
+    """Visual communication repository fixture."""
+    repo = VisualCommunicationRepository()
+    repo.connect()
+    yield repo
+    repo.close()

shared/repositories/tests/integration/image_repository_test.py

@@ -0,0 +1,156 @@
+"""Integration tests for ImageRepository class."""
+
+import numpy as np
+import pytest
+from PIL import Image
+
+from shared.repositories import ImageData, ImageRepository
+
+
+def same_image(
+    img_a: Image.Image,
+    img_b: Image.Image,
+) -> bool:
+    """Check if two images contain the same data."""
+    assert isinstance(img_a, Image.Image)
+    assert isinstance(img_b, Image.Image)
+    # check if images have a comparable number of channels
+    if img_a.getbands() != img_b.getbands():
+        return False
+    # calculate pixel difference between images
+    img_a_arr = np.asarray(img_a)
+    img_b_arr = np.asarray(img_b)
+    diff = np.subtract(img_a_arr, img_b_arr)
+    if np.sum(diff) != 0:
+        return False
+    return True
+
+
+def same_image_data(
+    data_a: ImageData,
+    data_b: ImageData,
+) -> bool:
+    """Check if two ImageData-objects contain the same data."""
+    assert isinstance(data_a, ImageData)
+    assert isinstance(data_b, ImageData)
+    # compare names
+    if data_a.name != data_b.name:
+        return False
+    # compare images
+    if not same_image(data_a.image, data_b.image):
+        return False
+    return True
+
+
+def test_should_have_context_handler():
+    """Test that class has a working context handler implemented."""
+    # ACT
+    with ImageRepository() as repo:
+        # ASSERT
+        assert repo.connected()
+
+
+def test_should_get_image_data(
+    image_repo: ImageRepository,
+    image_data_in_minio: ImageData,
+):
+    """Test getting image data."""
+    # ARRANGE
+    image_name = image_data_in_minio.name
+    # ACT
+    received_image_data = image_repo.get_data(image_name)
+    # ASSERT
+    assert received_image_data is not None
+    assert isinstance(received_image_data, ImageData)
+    assert same_image_data(image_data_in_minio, received_image_data)
+
+
+def test_should_get_none_when_no_image_data(
+    image_repo: ImageRepository,
+    image_data: ImageData,
+):
+    """Test getting None when no data is available."""
+    # ARRANGE
+    image_name = image_data.name
+    # ACT
+    received_image_data = image_repo.get_data(image_name)
+    # ASSERT
+    assert received_image_data is None
+
+
+def test_should_delete_image_data(
+    image_repo: ImageRepository,
+    image_data_in_minio: ImageData,
+):
+    """Test deleting image data."""
+    # ARRANGE
+    image_name = image_data_in_minio.name
+    # ACT
+    image_repo.remove_data(image_name)
+    received_image_data = image_repo.get_data(image_name)
+    # ASSERT
+    assert received_image_data is None
+
+
+def test_should_put_image_data(
+    image_repo: ImageRepository,
+    image_data: ImageData,
+):
+    """Test putting image data."""
+    # ARRANGE
+    image_name = image_data.name
+    # ACT
+    image_repo.put_data(image_data)
+    received_image_data = image_repo.get_data(image_name)
+    # ASSERT
+    assert received_image_data is not None
+    assert same_image_data(image_data, received_image_data)
+
+
+def test_should_update_image_data(
+    image_repo: ImageRepository,
+    image_data_in_minio: ImageData,
+):
+    """Test updating image data."""
+    # ARRANGE
+    updated_image_data = image_data_in_minio.model_copy(
+        update={
+            'image': Image.new(mode='RGB', size=(480, 480), color='white'),
+        },
+    )
+    image_name = updated_image_data.name
+    # ACT
+    image_repo.put_data(updated_image_data)
+    received_image_data = image_repo.get_data(image_name)
+    # ASSERT
+    assert not same_image_data(image_data_in_minio, updated_image_data)
+    assert received_image_data is not None
+    assert same_image_data(updated_image_data, received_image_data)
+
+
+def test_should_list_names(
+    image_repo: ImageRepository,
+    image_data_in_minio: ImageData,
+):
+    """Test get all image names."""
+    # ARRANGE
+    updated_image_data = image_data_in_minio.model_copy(
+        update={
+            'name': 'updated-test-image',
+        },
+    )
+    image_repo.put_data(updated_image_data)
+    expected_name_list = [
+        image_data_in_minio.name,
+        updated_image_data.name,
+    ]
+    # ACT
+    name_list = image_repo.list_names()
+    # ASSERT
+    assert len(name_list) == 2
+    for name in name_list:
+        assert name in expected_name_list
+
+
+if __name__ == '__main__':
+    pytest.main(['-s', '-v', __file__])

shared/repositories/tests/integration/minio_implementation_test.py

@@ -0,0 +1,133 @@
+"""Integration tests related to Minio implementation."""
+
+import logging
+import os
+from io import BytesIO
+from pathlib import Path
+
+import pytest
+
+from shared.repositories.src.implementations import MinioImplementation
+
+
+def same_data(
+    data_a: BytesIO,
+    data_b: BytesIO,
+) -> bool:
+    """Check if two BytesIO-objects contain the same data."""
+    assert isinstance(data_a, BytesIO)
+    assert isinstance(data_b, BytesIO)
+    # prepare for being read
+    data_a.seek(0)
+    data_b.seek(0)
+    # convert to bytes
+    data_a_bytes = data_a.read()
+    data_b_bytes = data_b.read()
+    # compare size
+    if len(data_a_bytes) != len(data_b_bytes):
+        logging.error(
+            'data has different length: %s and %s',
+            len(data_a_bytes),
+            len(data_b_bytes),
+        )
+        return False
+    # compare content
+    if data_a_bytes != data_b_bytes:
+        logging.error('data has different bytes')
+        return False
+    return True
+
+
+def test_should_connect_to_minio():
+    """Test connection to Minio."""
+    # ARRANGE
+    client = MinioImplementation()
+    # ACT
+    client.connect()
+    # ASSERT
+    assert client.connected()
+    client.close()
+
+
+def test_should_have_context_handler():
+    """Test that class has a working context handler implemented."""
+    # ACT
+    with MinioImplementation() as client:
+        # ASSERT
+        assert client.connected()
+
+
+def test_should_get_data(
+    minio_client: MinioImplementation,
+    buffer_in_minio: tuple[Path, BytesIO],
+):
+    """Test getting data from Minio."""
+    # ARRANGE
+    path, buffer = buffer_in_minio
+    # ACT
+    received_buffer = minio_client._get(path)
+    # ASSERT
+    assert received_buffer is not None
+    assert same_data(received_buffer, buffer)
+
+
+def test_should_get_none_when_no_data(
+    minio_client: MinioImplementation,
+):
+    """Test getting None when no data is available in Minio."""
+    # ARRANGE
+    nonexistent_path = Path('nonexistent-object-name')
+    # ACT
+    received_buffer = minio_client._get(nonexistent_path)
+    # ASSERT
+    assert received_buffer is None
+
+
+def test_should_delete_data(
+    minio_client: MinioImplementation,
+    buffer_in_minio: tuple[Path, BytesIO],
+):
+    """Test deleting data from Minio."""
+    # ARRANGE
+    path, _ = buffer_in_minio
+    # ACT
+    minio_client._delete(path)
+    # ASSERT
+    received_buffer = minio_client._get(path)
+    assert received_buffer is None
+
+
+def test_should_put_data(
+    minio_client: MinioImplementation,
+    buffer: BytesIO,
+):
+    """Test putting data in Minio."""
+    # ARRANGE
+    path = Path(os.getenv('MINIO_OBJECT_NAME', default=''))
+    # ACT
+    minio_client._put(path, buffer)
+    received_buffer = minio_client._get(path)
+    # ASSERT
+    assert received_buffer is not None
+    assert same_data(received_buffer, buffer)
+
+
+def test_should_update_data(
+    minio_client: MinioImplementation,
+    buffer_in_minio: tuple[Path, BytesIO],
+):
+    """Test updating data in Minio."""
+    # ARRANGE
+    path, buffer = buffer_in_minio
+    updated_buffer = BytesIO(buffer.getvalue() + b'extra data')
+    # ACT
+    minio_client._put(path, updated_buffer)
+    received_buffer = minio_client._get(path)
+    # ASSERT
+    assert not same_data(updated_buffer, buffer)
+    assert received_buffer is not None
+    assert same_data(received_buffer, updated_buffer)
+
+
+if __name__ == '__main__':
+    pytest.main(['-s', '-v', __file__])

shared/repositories/tests/integration/model_repository_test.py

@@ -0,0 +1,136 @@
+"""Integration tests for ModelRepository class."""
+
+from io import BytesIO
+
+import pytest
+
+from shared.repositories import ModelData, ModelRepository
+
+
+def same_buffer(
+    buffer_a: BytesIO,
+    buffer_b: BytesIO,
+) -> bool:
+    """Check if 2 buffers contain the same data."""
+    assert isinstance(buffer_a, BytesIO)
+    assert isinstance(buffer_b, BytesIO)
+    # read buffers
+    a_values = buffer_a.getvalue()
+    b_values = buffer_b.getvalue()
+    # compare length of buffers
+    if len(a_values) != len(b_values):
+        return False
+    # compare content of buffers
+    if a_values != b_values:
+        return False
+    return True
+
+
+def same_model_data(
+    data_a: ModelData,
+    data_b: ModelData,
+) -> bool:
+    """Check if to ModelData-objects contain the same data."""
+    assert isinstance(data_a, ModelData)
+    assert isinstance(data_b, ModelData)
+    # compare names
+    if data_a.buffer_checksum != data_b.buffer_checksum:
+        return False
+    # compare buffer
+    if not same_buffer(data_a.buffer, data_b.buffer):
+        return False
+    return True
+
+
+def test_should_have_context_handler():
+    """Test that class has a working context handler implemented."""
+    # ACT
+    with ModelRepository() as repo:
+        # ASSERT
+        assert repo.connected()
+
+
+def test_should_get_model_data(
+    model_repo: ModelRepository,
+    model_data_in_minio: ModelData,
+):
+    """Test getting model data."""
+    # ARRANGE
+    object_name = ModelRepository._build_object_name(model_data_in_minio)
+    # ACT
+    received_model_data = model_repo.get_data(object_name)
+    # ASSERT
+    assert received_model_data is not None
+    assert isinstance(received_model_data, ModelData)
+    assert same_model_data(model_data_in_minio, received_model_data)
+
+
+def test_should_get_none_when_no_model_data(
+    model_repo: ModelRepository,
+    model_data: ModelData,
+):
+    """Test getting None whne no data is available."""
+    # ARRANGE
+    object_name = ModelRepository._build_object_name(model_data)
+    # ACT
+    received_model_data = model_repo.get_data(object_name)
+    # ASSERT
+    assert received_model_data is None
+
+
+def test_should_delete_model_data(
+    model_repo: ModelRepository,
+    model_data_in_minio: ModelData,
+):
+    """Test deleting model data."""
+    # ARRANGE
+    object_name = ModelRepository._build_object_name(model_data_in_minio)
+    # ACT
+    model_repo.remove_data(object_name)
+    received_model_data = model_repo.get_data(object_name)
+    # ASSERT
+    assert received_model_data is None
+
+
+def test_should_put_model_data(
+    model_repo: ModelRepository,
+    model_data: ModelData,
+):
+    """Test putting model data."""
+    # ARRANGE
+    object_name = ModelRepository._build_object_name(model_data)
+    # ACT
+    model_repo.put_data(model_data)
+    received_model_data = model_repo.get_data(object_name)
+    # ASSERT
+    assert received_model_data is not None
+    assert same_model_data(model_data, received_model_data)
+
+
+def test_should_list_names(
+    model_repo: ModelRepository,
+    model_data_in_minio: ModelData,
+):
+    """Test get all model names."""
+    # ARRANGE
+    new_buffer_checksum = ModelData.calculate_checksum(model_data_in_minio.buffer)
+    updated_model_data = model_data_in_minio.model_copy(
+        update={
+            'buffer_checksum': new_buffer_checksum,
+        },
+    )
+    model_repo.put_data(updated_model_data)
+    expected_name_list = [
+        ModelRepository._build_object_name(model_data_in_minio),
+        ModelRepository._build_object_name(updated_model_data),
+    ]
+    # ACT
+    name_list = model_repo.list_names()
+    # ASSERT
+    assert len(name_list) == 2
+    for name in name_list:
+        assert name in expected_name_list
+
+
+if __name__ == '__main__':
+    pytest.main(['-s', '-v', __file__])