build process container

Author: catwhiskers

image-build-process/Dockerfile

@@ -3,36 +3,13 @@ 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}"
+COPY ocr_data_generator/ /opt/program/ocr_data_generator
 
-# RUN pip3 install --upgrade pip
+WORKDIR /opt/program
+RUN cd ./ocr_data_generator && pip install -r requirements.txt 
 
-# ## 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"]
 
 
 

image-build-process/ocr_data_generator/requirements.txt

@@ -0,0 +1,16 @@
+numpy==1.16.4
+opencv-python==4.3.0.38
+imutils==0.5.3
+trdg
+scipy==1.3.1
+glog
+matplotlib
+zxing
+fitz
+PyMuPDF
+fontTools
+tqdm==4.42.1
+shapely
+imgaug
+pyclipper
+lmdb