Merge pull request #586 from awslabs/batch-s3

feat(be): Batch on s3 processing

source/containers/document-pii-detection/Dockerfile

@@ -1,27 +1,30 @@
-FROM public.ecr.aws/lambda/python:3.9
+# temp stage
+FROM python:3.12-slim-bullseye as builder
 
-ARG FUNCTION_DIR="/opt/ml/code"
-COPY requirements.txt ${FUNCTION_DIR}/requirements.txt
-RUN python3.9 -m pip install -r ${FUNCTION_DIR}/requirements.txt
+# Install build dependencies
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends gcc g++
 
-COPY main.py parser_factory.py ${FUNCTION_DIR}/
-COPY parsers/ ${FUNCTION_DIR}/parsers/
+# Install Python dependencies first
+COPY requirements.txt .
+RUN pip wheel --no-cache-dir --no-deps --wheel-dir /app/wheels -r requirements.txt
 
-ARG OCR_MODEL_URL="https://aws-gcr-solutions-assets.s3.cn-northwest-1.amazonaws.com.cn/ai-solution-kit/infer-ocr-model/standard"
-ARG OCR_MODEL_VERSION="v1.0.0"
-ARG FD_MODEL_URL="https://aws-gcr-solutions-assets.s3.cn-northwest-1.amazonaws.com.cn/ai-solution-kit/face-detection"
-ARG FD_MODEL_VERSION="1.2.0"
+# prod stage
+FROM python:3.12-slim-bullseye
 
-RUN yum install -y wget
-RUN mkdir -p ${FUNCTION_DIR}/ocr_model
-RUN wget -c $OCR_MODEL_URL/$OCR_MODEL_VERSION/classifier.onnx -O ${FUNCTION_DIR}/ocr_model/classifier.onnx
-RUN wget -c $OCR_MODEL_URL/$OCR_MODEL_VERSION/det_standard.onnx -O ${FUNCTION_DIR}/ocr_model/det_standard.onnx
-RUN wget -c $OCR_MODEL_URL/$OCR_MODEL_VERSION/keys_v1.txt -O ${FUNCTION_DIR}/ocr_model/keys_v1.txt
-RUN wget -c $OCR_MODEL_URL/$OCR_MODEL_VERSION/rec_standard.onnx -O ${FUNCTION_DIR}/ocr_model/rec_standard.onnx
-RUN mkdir -p ${FUNCTION_DIR}/fd_model
-RUN wget -c ${FD_MODEL_URL}/${FD_MODEL_VERSION}/det.onnx -O ${FUNCTION_DIR}/fd_model/det.onnx
+ARG FUNCTION_DIR="/opt/ml/code"
 
 WORKDIR ${FUNCTION_DIR}
 
-# Command can be overwritten by providing a different command in the template directly.
-ENTRYPOINT ["python"]
+RUN apt-get update && apt-get install -y --no-install-recommends libmagic1 && rm -rf /var/lib/apt/lists/*
+
+COPY ocr_model/ ${FUNCTION_DIR}/ocr_model/
+COPY fd_model/ ${FUNCTION_DIR}/fd_model/
+COPY --from=builder /app/wheels ${FUNCTION_DIR}/wheels
+RUN pip install --no-cache ${FUNCTION_DIR}/wheels/*
+
+COPY parsers/ ${FUNCTION_DIR}/parsers/
+COPY requirements.txt main.py parser_factory.py ${FUNCTION_DIR}/
+
+
+CMD ["python3", "main.py"]