Creating a Custom Training Image (PyTorch + CPU/GPU)

Scenario

In this example, write a Dockerfile to create a custom image on a Linux x86_64 server running Ubuntu 18.04.

Objective: Build and install container images of the following software and use the images and CPUs/GPUs for training jobs on ModelArts.

• ubuntu-18.04
• cuda-11.1
• python-3.7.13
• pytorch-1.8.1

Procedure

Before using a custom image to create a training job, get familiar with Docker and have development experience. The following is the detailed procedure:

1. Prerequisites
2. Step 1 Creating an OBS Bucket and Folder
3. Step 2 Preparing the Training Script and Uploading It to OBS
4. Step 3 Preparing a Host
5. Step 4 Creating a Custom Image
6. Step 5 Uploading an Image to SWR
7. Step 6 Creating a Training Job on ModelArts

Prerequisites

You have registered a Huawei ID and enabled Huawei Cloud services, and the account is not in arrears or frozen.

Step 1 Creating an OBS Bucket and Folder

Create a bucket and the folders in OBS for storing the sample dataset and training code. Table 1 lists the folders to be created. Replace the bucket name test-modelarts and folder names in the example with actual names.

For details about how to create an OBS bucket and folder, see Creating a Bucket and Creating a Folder.

Ensure that the OBS directory you use and ModelArts are in the same region.

Step 2 Preparing the Training Script and Uploading It to OBS

Prepare the training script pytorch-verification.py and upload it to the obs://test-modelarts/pytorch/demo-code/ folder of the OBS bucket.

The pytorch-verification.py file contains the following information:

import torch
import torch.nn as nn

x = torch.randn(5, 3)
print(x)

available_dev = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
y = torch.randn(5, 3).to(available_dev)
print(y)

Step 3 Preparing a Host

Obtain a Linux x86_64 server running Ubuntu 18.04. Either an ECS or your local PC will do.

For details about how to purchase an ECS, see Purchasing and Logging In to a Linux ECS. Set CPU Architecture to x86 and Image to Public image. Ubuntu 18.04 images are recommended.

Step 4 Creating a Custom Image

Create a container image with the following configurations and use the image to create a training job on ModelArts:

• ubuntu-18.04
• cuda-11.1
• python-3.7.13
• pytorch-1.8.1

This section describes how to write a Dockerfile to create a custom image.

1. Install Docker.

   The following uses the Linux x86_64 OS as an example to describe how to obtain the Docker installation package. For details about how to install Docker, see official Docker documents. Run the following commands to install Docker:

   curl -fsSL get.docker.com -o get-docker.sh
   sh get-docker.sh

   If the docker images command is executed, Docker has been installed. In this case, skip this step.

2. Run the following command to check the Docker Engine version:

   docker version | grep -A 1 Engine

   The command output is as follows:

   ...
   Engine:
     Version:          18.09.0

   

   Use the Docker engine of the preceding version or later to create a custom image.

3. Create a folder named context.

   mkdir -p context

4. Obtain the pip.conf file. In this example, the pip source provided by Huawei open-source image site is used. The content of pip.conf is as follows:

   [global]
   index-url = https://repo.huaweiexamplecloud.com/repository/pypi/simple
   trusted-host = repo.huaweiexamplecloud.com
   timeout = 120

   

   To obtain pip.conf, go to Huawei Mirrors at https://mirrors.huaweicloud.com/home and search for pypi.

5. Download the torch*.whl file.

   Download the following .whl files from https://download.pytorch.org/whl/torch_stable.html:

   • torch-1.8.1+cu111-cp37-cp37m-linux_x86_64.whl
   • torchaudio-0.8.1-cp37-cp37m-linux_x86_64.whl
   • torchvision-0.9.1+cu111-cp37-cp37m-linux_x86_64.whl
   

   The plus sign (+) must be URL-encoded to %2B. When searching for the target file name on the preceding website, replace the plus sign (+) in the original file name with %2B.

   Example: torch-1.8.1%2Bcu111-cp37-cp37m-linux_x86_64.whl

6. Download the Miniconda3 installation file.

   Download the Miniconda3 py37 4.12.0 installation file (Python 3.7.13) from https://repo.anaconda.com/miniconda/Miniconda3-py37_4.12.0-Linux-x86_64.sh.

7. Store the pip source file, torch*.whl file, and Miniconda3 installation file in the context folder, which is as follows:

   context
   ├── Miniconda3-py37_4.12.0-Linux-x86_64.sh
   ├── pip.conf
   ├── torch-1.8.1+cu111-cp37-cp37m-linux_x86_64.whl
   ├── torchaudio-0.8.1-cp37-cp37m-linux_x86_64.whl
   └── torchvision-0.9.1+cu111-cp37-cp37m-linux_x86_64.whl

8. Write the Dockerfile of the container image.
   Create an empty file named Dockerfile in the context folder and copy the following content to the file:

   # The server on which the container image is created must access the Internet.
   
   # Base container image at https://github.com/NVIDIA/nvidia-docker/wiki/CUDA
   # 
   # https://docs.docker.com/develop/develop-images/multistage-build/#use-multi-stage-builds
   # require Docker Engine >= 17.05
   #
   # builder stage
   FROM nvidia/cuda:11.1.1-runtime-ubuntu18.04 AS builder
   
   # The default user of the base container image is root.
   # USER root
   
   # Use the PyPI configuration provided by the open-source image site.
   RUN mkdir -p /root/.pip/
   COPY pip.conf /root/.pip/pip.conf
   
   # Copy the installation files to the /tmp directory in the base container image.
   COPY Miniconda3-py37_4.12.0-Linux-x86_64.sh /tmp
   COPY torch-1.8.1+cu111-cp37-cp37m-linux_x86_64.whl /tmp
   COPY torchvision-0.9.1+cu111-cp37-cp37m-linux_x86_64.whl /tmp
   COPY torchaudio-0.8.1-cp37-cp37m-linux_x86_64.whl /tmp
   
   # https://conda.io/projects/conda/en/latest/user-guide/install/linux.html#installing-on-linux
   # Install Miniconda3 in the /home/ma-user/miniconda3 directory of the base container image.
   RUN bash /tmp/Miniconda3-py37_4.12.0-Linux-x86_64.sh -b -p /home/ma-user/miniconda3
   
   # Install torch*.whl using the default Miniconda3 Python environment in /home/ma-user/miniconda3/bin/pip.
   RUN cd /tmp && \
       /home/ma-user/miniconda3/bin/pip install --no-cache-dir \
       /tmp/torch-1.8.1+cu111-cp37-cp37m-linux_x86_64.whl \
       /tmp/torchvision-0.9.1+cu111-cp37-cp37m-linux_x86_64.whl \
       /tmp/torchaudio-0.8.1-cp37-cp37m-linux_x86_64.whl
   
   # Create the container image.
   FROM nvidia/cuda:11.1.1-runtime-ubuntu18.04
   
   # Install vim and cURL in the open-source image site.
   RUN cp -a /etc/apt/sources.list /etc/apt/sources.list.bak && \
       sed -i "s@http://.*archive.ubuntu.com@http://repo.huaweiexamplecloud.com@g" /etc/apt/sources.list && \
       sed -i "s@http://.*security.ubuntu.com@http://repo.huaweiexamplecloud.com@g" /etc/apt/sources.list && \
       apt-get update && \
       apt-get install -y vim curl && \
       apt-get clean && \
       mv /etc/apt/sources.list.bak /etc/apt/sources.list
   
   # Add user ma-user (UID = 1000, GID = 100).
   # A user group whose GID is 100 exists in the base container image. User ma-user can directly run the following command:
   RUN useradd -m -d /home/ma-user -s /bin/bash -g 100 -u 1000 ma-user
   
   # Copy the /home/ma-user/miniconda3 directory from the builder stage to the directory with the same name in the current container image.
   COPY --chown=ma-user:100 --from=builder /home/ma-user/miniconda3 /home/ma-user/miniconda3
   
   # Configure the preset environment variables of the container image.
   # Set PYTHONUNBUFFERED to 1 to prevent log loss.
   ENV PATH=$PATH:/home/ma-user/miniconda3/bin \
       PYTHONUNBUFFERED=1
   
   # Configure the default user and working directory of the container image.
   USER ma-user
   WORKDIR /home/ma-user

   For details about how to write a Dockerfile, see official Docker documents.

9. Ensure that the Dockerfile has been created. The following shows the context folder:

   context
   ├── Dockerfile
   ├── Miniconda3-py37_4.12.0-Linux-x86_64.sh
   ├── pip.conf
   ├── torch-1.8.1+cu111-cp37-cp37m-linux_x86_64.whl
   ├── torchaudio-0.8.1-cp37-cp37m-linux_x86_64.whl
   └── torchvision-0.9.1+cu111-cp37-cp37m-linux_x86_64.whl

10. Create the container image. Run the following command in the directory where the Dockerfile is stored to build the container image pytorch:1.8.1-cuda11.1:

    1	docker build . -t pytorch:1.8.1-cuda11.1

    The following log shows that the image has been created.

    Successfully tagged pytorch:1.8.1-cuda11.1

Step 5 Uploading an Image to SWR

1. Log in to the SWR console and select a region. It must share the same region with ModelArts. Otherwise, the image cannot be selected.
2. Click Create Organization in the upper right corner and enter an organization name to create an organization. Customize the organization name. Replace the organization name deep-learning in subsequent commands with the actual organization name.
3. Click Generate Login Command in the upper right corner to obtain the login command. In this example, the temporary login command is copied.
4. Log in to the local environment as user root and enter the copied temporary login command.
5. Upload the image to SWR.
   1. Tag the uploaded image.

      # Replace the region, domain, as well as organization name deep-learning with the actual values.
      sudo docker tag pytorch:1.8.1-cuda11.1 swr.{region-id}.{domain}/deep-learning/pytorch:1.8.1-cuda11.1

   2. Upload the image.

      # Replace the region, domain, as well as organization name deep-learning with the actual values.
      sudo docker push swr.{region-id}.{domain}/deep-learning/pytorch:1.8.1-cuda11.1

6. After the image is uploaded, choose My Images in navigation pane on the left of the SWR console to view the uploaded custom images.

Step 6 Creating a Training Job on ModelArts

1. Log in to the ModelArts console and check whether access authorization has been configured for your account. If not, configure the authorization by referring to Configuring Agency Authorization for ModelArts with One Click. For users who used AKs for authorization, you are advised to clear the existing authorization and use an agency.
2. In the navigation pane, choose Model Training > Training Jobs.
3. On the Create Training Job page, configure parameters and click Submit.
   • Algorithm Type: Custom algorithm
   • Boot Mode: Custom image
   • Image path: image created in Step 5 Uploading an Image to SWR
   • Code Directory: directory where the boot script file is stored in OBS, for example, obs://test-modelarts/pytorch/demo-code/. The training code is automatically downloaded to the ${MA_JOB_DIR}/demo-code directory of the training container. demo-code (customizable) is the last-level directory of the OBS path.
   • Boot Command: /home/ma-user/miniconda3/bin/python ${MA_JOB_DIR}/demo-code/pytorch-verification.py. demo-code (customizable) is the last-level directory of the OBS path.
   • Resource Pool: Public resource pools
   • Resource Type: Select GPU or CPU.
   • Persistent Log Saving: enabled
   • Job Log Path: OBS path to stored training logs, for example, obs://test-modelarts/pytorch/log/.
4. Confirm the configurations of the training job and click Submit.
5. Wait until the training job is created.

   After you submit the job creation request, the system will automatically perform operations on the backend, such as downloading the container image and code directory and running the boot command. A training job requires a certain period of time for running. The duration ranges from dozens of minutes to several hours, depending on the service logic and selected resources. After the training job is executed, the log similar to the following is output.

   Figure 1 Run logs of training jobs with GPU specifications