update

Dockerfile

@@ -12,55 +12,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-ARG FROM_IMAGE_NAME=nvcr.io/nvidia/tensorflow:22.09-tf1-py3
-FROM ${FROM_IMAGE_NAME}
+FROM=image.sourcefind.cn:5000/dcu/admin/base/custom:mlperf-minigo-latest
 
-# Install system dependencies
-RUN apt-get update \
- && apt-get install -y --no-install-recommends \
-        cmake \
-        python2.7 \
-        infiniband-diags \
-        pciutils \
- && rm -rf /var/lib/apt/lists/*
 
-# Copy MiniGo code
-WORKDIR /opt/reinforcement
-COPY . .
-
-ENV BOARD_SIZE 19
-ENV MINIGO_BAZEL_CACHE_DIR /opt/reinforcement/minigo-bazel-cache
-
-# Copy TF dependency
-RUN mkdir minigo/cc/tensorflow/lib \
- && cp /usr/local/lib/python3.8/dist-packages/tensorflow_core/libtensorflow_framework.so.1 minigo/cc/tensorflow/lib \
- && cp /usr/local/lib/python3.8/dist-packages/tensorflow_core/libtensorflow_cc.so.1 minigo/cc/tensorflow/lib \
- && cp -r /usr/local/lib/python3.8/dist-packages/tensorflow_core/include minigo/cc/tensorflow/include
-
-# Install Python dependencies
-WORKDIR /opt/reinforcement/minigo
-RUN pip3 install --no-cache-dir git+https://github.com/mlcommons/logging.git \
- && pip3 install --no-cache-dir -r requirements.txt
-
-# Install pybind11 to enable C++-python interface
-RUN pip3 install --no-cache-dir pytest \
- && git clone --branch v2.4.3 --depth 1 https://github.com/pybind/pybind11 /usr/local/src/pybind11 \
- && cd /usr/local/src/pybind11 \
- && cmake . \
- && make install -j$(nproc) \
- && pip3 install --no-cache-dir .
-
-# Build Minigo
-RUN mkdir -p "${MINIGO_BAZEL_CACHE_DIR}" \
- && bazel --output_user_root="${MINIGO_BAZEL_CACHE_DIR}" build -c opt \
-      --cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0" \
-      --copt=-O3 \
-      --define=board_size="${BOARD_SIZE}" \
-      --define=tf=1 \
-      cc:minigo_python.so
-
-ENV PYTHONPATH "${PYTHONPATH}:/opt/reinforcement/minigo/bazel-bin/cc"
-RUN echo '/usr/local/lib/python3.8/dist-packages/tensorflow_core' > /etc/ld.so.conf.d/tensorflow.conf && ldconfig
-
-# back to where run* files are
-WORKDIR /opt/reinforcement

README.md

@@ -19,6 +19,8 @@ Minogo是一个基于深度强化学习的围棋程序，模型灵感来源于Go
 
 ## 环境配置
 
+**Docker (方法一)**
+
 提供[光源](https://www.sourcefind.cn/#/service-details)拉取的训练的docker镜像：
 
     docker pull image.sourcefind.cn:5000/dcu/admin/base/custom:mlperf-minigo-latest
@@ -27,6 +29,14 @@ Minogo是一个基于深度强化学习的围棋程序，模型灵感来源于Go
     # <Container Path>容器映射路径
     docker run -it --name mlperf_minigo --shm-size=32G  --device=/dev/kfd --device=/dev/dri/ --cap-add=SYS_PTRACE --security-opt seccomp=unconfined --ulimit memlock=-1:-1 --ipc=host --network host --group-add video -v <Host Path>:<Container Path> <Image ID> /bin/bash
 
+**Dockerfile (方法二)**
+
+    docker build --no-cache -t mlperf_minigo:latest
+    docker run -it --name mlperf_minigo --shm-size=32G  --device=/dev/kfd --device=/dev/dri/ --cap-add=SYS_PTRACE --security-opt seccomp=unconfined --ulimit memlock=-1:-1 --ipc=host --network host --group-add video -v <Host Path>:<Container Path> <Image ID> /bin/bash
+    # <Image ID>用上面拉取docker镜像的ID替换
+    # <Host Path>主机端路径
+    # <Container Path>容器映射路径
+
 镜像版本依赖：
 
 * DTK驱动：dtk22.04.2
@@ -85,7 +95,7 @@ Minogo是一个基于深度强化学习的围棋程序，模型灵感来源于Go
 
 ## 源码仓库及问题反馈
 * https://developer.hpccube.com/codes/modelzoo/mlperf_minigo_tensorflow
-## 参考
+## 参考资料
 * https://mlcommons.org/en/
 * https://github.com/mlcommons
 * https://github.com/mlcommons/training_results_v2.1/tree/main/NVIDIA/benchmarks/minigo/implementations/tensorflow-22.09