Docker Image for RAG applications

nlp-notebook/cuda12-pytorch-2.2.2-RAG/Dockerfile

@@ -0,0 +1,31 @@
+ARG BASE_IMAGE=quay.io/a2s-institute/ml-notebook:cuda12-pytorch-2.2.2
+
+FROM $BASE_IMAGE
+
+LABEL maintainer="Mohammad Wasil <e2x@inf.h-brs.de>"
+
+USER $NB_USER
+
+COPY requirements.txt /tmp/requirements.txt
+COPY chroma_fix.sh /tmp/chroma_fix.sh
+RUN pip install --upgrade pip && \
+    pip install --no-cache-dir -r /tmp/requirements.txt
+
+
+
+USER root
+
+RUN curl -fsSL https://ollama.com/install.sh | sh
+
+RUN chmod +x /tmp/chroma_fix.sh && /tmp/chroma_fix.sh
+
+# clean up
+RUN rm /tmp/requirements.txt && \
+    /opt/conda/bin/conda clean -afy && \
+    find /opt/conda/ -follow -type f -name '*.a' -delete && \
+    find /opt/conda/ -follow -type f -name '*.pyc' -delete && \
+    find /opt/conda/ -follow -type f -name '*.js.map' -delete
+
+USER $NB_USER
+
+CMD ["ollama","serve"]

nlp-notebook/cuda12-pytorch-2.2.2-RAG/chroma_fix.sh

@@ -0,0 +1,26 @@
+#!/bin/bash
+
+# File you want to prepend lines to
+TARGET_FILE="./opt/conda/lib/python3.11/site-packages/chromadb/__init__.py"
+
+# Temporary file to store the existing content
+TEMP_FILE=$(mktemp)
+
+# Check if the file exists
+if [ -f "$TARGET_FILE" ]; then
+  # Write the new lines to the temporary file
+  echo "__import__('pysqlite3')" > "$TEMP_FILE"
+  echo "import sys" >> "$TEMP_FILE"
+  echo "sys.modules['sqlite3'] = sys.modules.pop('pysqlite3')" >> "$TEMP_FILE"
+
+  # Append the original content to the temporary file
+  cat "$TARGET_FILE" >> "$TEMP_FILE"
+
+  # Move the temporary file to the original file
+  mv "$TEMP_FILE" "$TARGET_FILE"
+
+  echo "Lines have been prepended to $TARGET_FILE"
+else
+  echo "File not found!"
+  rm "$TEMP_FILE"
+fi

nlp-notebook/cuda12-pytorch-2.2.2-RAG/requirements.txt

@@ -0,0 +1,13 @@
+llama-index-core==0.10.63
+llama-index-embeddings-huggingface==0.2.2
+llama-index-llms-ollama==0.2.2
+llama-index-readers-file==0.1.32
+llama-index-vector-stores-chroma==0.1.10
+chromadb==0.5.3
+pysqlite3-binary==0.5.3
+PyMuPDF==1.24.9
+spacy
+nltk
+gensim
+wikipedia-api
+llama-cpp-python