diff --git a/ptuning/docker/guide.sh b/ptuning/docker/guide.sh
new file mode 100644
index 0000000..cd7ba5c
--- /dev/null
+++ b/ptuning/docker/guide.sh
@@ -0,0 +1,3 @@
+#!/usr/bin/env bash
+
+docker build 
diff --git a/ptuning/docker/train_gpu/start.sh b/ptuning/docker/train_gpu/start.sh
new file mode 100644
index 0000000..3269245
--- /dev/null
+++ b/ptuning/docker/train_gpu/start.sh
@@ -0,0 +1,12 @@
+#!/usr/bin/env bash
+
+# activate base
+source /app/dev/miniconda3/bin/activate base
+conda env list
+
+# update source
+cd /app/source/ChatGLM-6B/ptuning || exit 1
+git pull
+
+# start training
+bash train.sh 1>log_train.log 2>&1 &
diff --git a/ptuning/docker/train_gpu/train.gpu.Dockerfile b/ptuning/docker/train_gpu/train.gpu.Dockerfile
new file mode 100644
index 0000000..2f285a6
--- /dev/null
+++ b/ptuning/docker/train_gpu/train.gpu.Dockerfile
@@ -0,0 +1,35 @@
+# nvidia/cuda: https://hub.docker.com/r/nvidia/cuda/tags?page=1&name=devel-ubuntu18.04
+# 1. check host nvidia drvier version: nvidia-smi
+#           Nvidia driver version compatible cuda version: https://docs.nvidia.com/deploy/cuda-compatibility/#minor-version-compatibility
+#                   CUDA 12.x: Linux x86_64 minimum driver version >= 525.60.13
+#                   CUDA 11.x: Linux x86_64 minimum driver version >= 450.80.02
+# 2. check target PyTorch version: import torch; print(torch.cuda.is_available()) -> True
+#           PyTorch version compatible cuda version: https://pytorch.org/get-started/previous-versions/
+FROM nvidia/cuda:11.2.2-devel-ubuntu18.04
+
+# set timezone
+RUN apt-get update -y && apt-get install -y tzdata
+ENV TZ=Asia/Shanghai
+
+# install miniconda: https://docs.conda.io/en/latest/miniconda.html#linux-installers
+RUN apt-get -y update && apt-get install -y wget && \
+    mkdir -p /app && wget -P /app https://repo.anaconda.com/miniconda/Miniconda3-py38_23.1.0-1-Linux-x86_64.sh && \
+    cd /app && bash Miniconda3-py38_23.1.0-1-Linux-x86_64.sh -b -p /app/dev/miniconda3 && \
+    source /app/dev/miniconda3/bin/activate base && conda init bash && source ~/.bashrc && conda env list
+
+# install chatglm-6b dependencies
+RUN source /app/dev/miniconda3/bin/activate base && conda env list && \
+    pip install protobuf \
+                torch transformers datasets accelerate peft \
+                icetk cpm_kernels rouge_chinese nltk jieba \
+                gradio==3.20.0 mdtex2html fastapi uvicorn requests
+RUN mkdir -p /app/source && cd /app/source && \
+    apt-get update -y && apt-get install -y git && \
+    git clone --recursive https://github.com/zealotpb/ChatGLM-6B.git
+
+# copy start.sh
+COPY start.sh /usr/bin/start
+RUN chmod a+x /usr/bin/start
+
+# install necessary debug tools
+RUN apt-get update -y && apt-get install -y vim htop
diff --git a/ptuning/docker/train_gpu/train.sh b/ptuning/docker/train_gpu/train.sh
new file mode 100644
index 0000000..69ecd52
--- /dev/null
+++ b/ptuning/docker/train_gpu/train.sh
@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+
+# activate base
+source /app/dev/miniconda3/bin/activate base
+conda env list