add produce training data dockerfile

Author: catwhiskers

image-build-process/Dockerfile

@@ -0,0 +1,38 @@
+FROM python:3.7
+
+
+RUN apt-get update
+RUN apt-get install ffmpeg libsm6 libxext6  -y
+# ENV LANG=en_US.utf8
+# ENV LANG=C.UTF-8
+
+# ENV PYTHONUNBUFFERED=TRUE
+# ENV PYTHONDONTWRITEBYTECODE=TRUE
+# ENV PATH="/opt/program:${PATH}"
+
+# RUN pip3 install --upgrade pip
+
+# ## install paddlepaddle framework
+# RUN pip3 install paddlepaddle-gpu -i https://mirror.baidu.com/pypi/simple
+# RUN pip3 install paddleocr==2.0.1
+
+# ## clone PaddleOCR source code 
+# RUN git clone -b release/2.1 https://github.com/PaddlePaddle/PaddleOCR.git /opt/program/
+
+
+# #download pretrained model for finetunine
+# RUN mkdir /opt/program/pretrain/
+# RUN cd /opt/program/pretrain/
+# RUN wget -P /opt/program/pretrain/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_train.tar && tar -xf /opt/program/pretrain/ch_ppocr_mobile_v2.0_rec_train.tar -C /opt/program/pretrain/ && rm -rf /opt/program/pretrain/ch_ppocr_mobile_v2.0_rec_train.tar
+
+# Set up the program in the image
+RUN git clone https://github.com/catwhiskers/ocr_data_generator.git
+
+#WORKDIR /opt/program
+RUN cd ./ocr_data_generator && pip install -r requirements.txt && python generate_data.py seto_font train data/ZH_1.txt setofont 
+
+
+# ENTRYPOINT ["python3", "train.py"]
+
+
+