Revert and add Cmake
Browse files- Dockerfile +7 -11
Dockerfile
CHANGED
@@ -18,22 +18,18 @@ RUN useradd -m -u 1000 user
|
|
18 |
# Set the working directory
|
19 |
WORKDIR /app
|
20 |
|
21 |
-
# Create and activate virtual environment
|
22 |
-
RUN python3 -m venv /app/venv
|
23 |
-
ENV PATH="/app/venv/bin:$PATH"
|
24 |
-
|
25 |
# Copy the requirements file
|
26 |
COPY requirements.txt .
|
27 |
|
28 |
# Install Python dependencies
|
29 |
-
RUN
|
30 |
-
|
31 |
-
pip install --no-cache-dir -r requirements.txt
|
32 |
|
33 |
# Install llama-cpp-python with CUDA support
|
34 |
-
|
35 |
-
|
36 |
-
|
|
|
37 |
|
38 |
# Copy the rest of the application code
|
39 |
COPY . .
|
@@ -48,4 +44,4 @@ USER user
|
|
48 |
ENV PATH="/home/user/.local/bin:$PATH"
|
49 |
|
50 |
# Run the application
|
51 |
-
CMD ["
|
|
|
18 |
# Set the working directory
|
19 |
WORKDIR /app
|
20 |
|
|
|
|
|
|
|
|
|
21 |
# Copy the requirements file
|
22 |
COPY requirements.txt .
|
23 |
|
24 |
# Install Python dependencies
|
25 |
+
RUN pip3 install --no-cache-dir --upgrade pip && \
|
26 |
+
pip3 install --no-cache-dir -r requirements.txt
|
|
|
27 |
|
28 |
# Install llama-cpp-python with CUDA support
|
29 |
+
ENV FORCE_CMAKE=1
|
30 |
+
ENV CMAKE_ARGS="-DLLAMA_CUBLAS=on"
|
31 |
+
RUN pip3 install llama-cpp-python \
|
32 |
+
--extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cu121
|
33 |
|
34 |
# Copy the rest of the application code
|
35 |
COPY . .
|
|
|
44 |
ENV PATH="/home/user/.local/bin:$PATH"
|
45 |
|
46 |
# Run the application
|
47 |
+
CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]
|