debug

Dockerfile

@@ -43,8 +43,14 @@ RUN python3 -m pip install --upgrade pip
 COPY ./requirements.txt /app/requirements.txt
 RUN pip3 install --break-system-packages --no-cache-dir --upgrade -r /app/requirements.txt
 RUN pip3 install --break-system-packages https://files.pythonhosted.org/packages/13/87/e0cb08c2d4bd7d38ab63816b306c8b1e7cfdc0e59bd54462e8b0df069078/semantic_text_splitter-0.6.3-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
-RUN pip3 install --break-system-packages llama_cpp_python
-RUN FORCE_CMAKE=1 CMAKE_SYSTEM_PROCESSOR=AMD64 pip3 install --break-system-packages --verbose --no-cache-dir llama-cpp-python --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu
+
+#RUN pip3 install --break-system-packages llama_cpp_python
+#RUN FORCE_CMAKE=1 CMAKE_SYSTEM_PROCESSOR=AMD64 pip3 install --break-system-packages --verbose --no-cache-dir llama-cpp-python --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu
+#RUN FORCE_CMAKE=1 CMAKE_SYSTEM_PROCESSOR=AMD64 pip3 install --break-system-packages --verbose --no-cache-dir llama-cpp-python 
+
+#RUN CMAKE_ARGS="-DLLAMA_CUBLAS=on -DCUDA_PATH=/usr/local/cuda-12.2 -DCUDAToolkit_ROOT=/usr/local/cuda-12.2 -DCUDAToolkit_INCLUDE_DIR=/usr/local/cuda-12/include -DCUDAToolkit_LIBRARY_DIR=/usr/local/cuda-12.2/lib64" FORCE_CMAKE=1 pip install llama-cpp-python - no-cache-dir
+RUN CMAKE_ARGS="-DLLAMA_CUBLAS=on FORCE_CMAKE=1 pip install --break-system-packages llama-cpp-python --no-cache-dir
+
 RUN pip3 install --break-system-packages cffi
 
 # Install text2vec-transformers

app.py

@@ -398,7 +398,7 @@ try:
         ###############################################################################
         # Initial the the sentence transformer and encode the query prompt.
         logger.debug(f"#### Encode text query prompt to create vectors. {promptText}")
-        model = SentenceTransformer('/app/text2vec-transformers/multi-qa-MiniLM-L6-cos-v1')
+        model = SentenceTransformer('/app/multi-qa-MiniLM-L6-cos-v1')
         vector = model.encode(promptText)
         
         logLevel = logger.getEffectiveLevel()
@@ -406,7 +406,6 @@ try:
             wrks = str(vector)
             logger.debug(f"### vector: {wrks}")
 
-        
         vectorList = []
         for vec in vector:
             vectorList.append(vec)

startup.sh

@@ -7,6 +7,8 @@
 #####################################
 exec &> /app/startup.log
 
+echo "#### pip3 list"; pip3 list
+
 stdbuf -o0 echo "#### startup.sh entered."
 #stdbuf -o0 echo "### ps -ef 1"; ps -ef