feat(download_model.py): remove download_model.py during build, it causing big image size

.gitignore

@@ -1 +1,2 @@
-.idea
+.idea
+dep_size.txt

Dockerfile

@@ -6,6 +6,8 @@ ENV PATH="/home/user/.local/bin:$PATH"
 
 WORKDIR /app
 
+# revent python from generating .pyc files you don't need
+ENV PYTHONDONTWRITEBYTECODE=1
 COPY --chown=user ./requirements.txt requirements.txt
 RUN pip install --no-cache-dir -r requirements.txt --extra-index-url https://download.pytorch.org/whl/cu113
 
@@ -23,7 +25,11 @@ COPY --chown=user . /app
 #
 # AFTER TRIAL AND ERROR WE GOT 16GB (16431849854 bytes) OF LAYERS :(
 #
-RUN --mount=type=secret,id=HF_TOKEN,mode=0444,required=true HF_TOKEN=$(cat /run/secrets/HF_TOKEN) python /app/download_model.py
+# RUN --mount=type=secret,id=HF_TOKEN,mode=0444,required=true HF_TOKEN=$(cat /run/secrets/HF_TOKEN) python /app/download_model.py
+
+# Getting the biggest dependencies
+# https://stackoverflow.com/a/78014074
+RUN python -c "for d in __import__('importlib.metadata').metadata.distributions(): print('{:>12.3f} KiB   {}'.format(sum(0 if not f.locate().is_file() else f.locate().stat().st_size for f in d.files) / 1024, d.name))" | sort > dep_sizes.txt
 
 EXPOSE 7860
 

openai_compatible_api_server.py

@@ -287,6 +287,16 @@ def engine_client(request: Request) -> EngineClient:
     return request.app.state.engine_client
 
 
+@router.get("/")
+async def home(raw_request: Request) -> Response:
+    dep_sizes: str = ""
+    with open('/app/dep_sizes.txt', mode='r') as f:
+        dep_sizes = f.readline()
+        f.close()
+
+    return Response(status_code=200, content=dep_sizes)
+
+
 @router.get("/health")
 async def health(raw_request: Request) -> Response:
     """Health check."""