Update Dockerfile

Dockerfile

@@ -26,6 +26,7 @@ RUN apt-get update && \
     nvidia-driver-550 \
     python3.10 \
     python3.10-venv \
+    python-is-python3 \
     ffmpeg
 
 ENV USER='user'    
@@ -46,18 +47,18 @@ COPY --chown=1000 . ${APPDIR}
 #     pyenv global ${PYTHON_VERSION} && \
 #     pyenv rehash && \
 
-RUN python3.10 -m venv .venv 
-RUN . .venv/bin/activate
-RUN pip install --no-cache-dir -U pip setuptools wheel && \
-    pip install "huggingface-hub" "hf-transfer" "gradio[oauth]>=4.28.0" "gradio_huggingfacehub_search==0.0.7" "APScheduler"
-RUN deactivate
+# RUN python3.10 -m venv .venv 
+# RUN . .venv/bin/activate
+RUN python pip install --no-cache-dir -U pip setuptools wheel
+RUN python pip install "huggingface-hub" "hf-transfer" "gradio[oauth]>=4.28.0" "gradio_huggingfacehub_search==0.0.7" "APScheduler"
+# RUN deactivate
 
 RUN git clone https://github.com/ollama/ollama 
 RUN git clone https://github.com/ggerganov/llama.cpp
 COPY groups_merged.txt ${HOME}/app/llama.cpp/.
 
-ENV PYTHONPATH=${APPDIR}/.venv/bin \
-    PYTHONUNBUFFERED=1 \
+# ENV PYTHONPATH=${APPDIR}/.venv/bin \
+ENV PYTHONUNBUFFERED=1 \
     HF_HUB_ENABLE_HF_TRANSFER=1 \
     GRADIO_ALLOW_FLAGGING=never \
     GRADIO_NUM_PORTS=1 \
@@ -77,12 +78,12 @@ RUN OLLAMA_CUSTOM_CPU_DEFS="-DGGML_AVX=on -DGGML_AVX2=on -DGGML_F16C=on -DGGML_F
     go install . --verbose
 
 WORKDIR ${APPDIR}/llama.cpp
-RUN pip install -r requirements.txt \
+RUN python pip install -r requirements.txt \
     LLAMA_CUDA=0 make -j llama-quantize --verbose
 
 WORKDIR ${APPDIR}
 
 # EXPOSE map[11434/tcp:{}]
 RUN ollama serve --verbose & sleep 5
-RUN . .venv/bin/activate
+# RUN . .venv/bin/activate
 ENTRYPOINT python app.py --verbose