{"id":23683840,"url":"https://github.com/aime-team/aime-mlc","last_synced_at":"2025-04-11T04:51:20.414Z","repository":{"id":270157653,"uuid":"909450113","full_name":"aime-team/aime-mlc","owner":"aime-team","description":"AIME Machine Learning Containers (MLC) management system 2.0","archived":false,"fork":false,"pushed_at":"2025-04-02T07:53:49.000Z","size":242,"stargazers_count":3,"open_issues_count":3,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-02T08:33:18.545Z","etag":null,"topics":["docker","pytorch","tensorflow"],"latest_commit_sha":null,"homepage":"https://www.aime.info/mlc","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/aime-team.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.txt","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-12-28T18:32:40.000Z","updated_at":"2025-04-02T07:53:53.000Z","dependencies_parsed_at":"2025-01-29T17:30:49.365Z","dependency_job_id":"38abf337-c63b-4800-a721-e3b432bb3b9f","html_url":"https://github.com/aime-team/aime-mlc","commit_stats":null,"previous_names":["aime-team/aime-mlc"],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aime-team%2Faime-mlc","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aime-team%2Faime-mlc/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aime-team%2Faime-mlc/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aime-team%2Faime-mlc/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/aime-team","download_url":"https://codeload.github.com/aime-team/aime-mlc/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248345289,"owners_count":21088243,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["docker","pytorch","tensorflow"],"created_at":"2024-12-29T20:21:34.013Z","updated_at":"2025-04-11T04:51:20.403Z","avatar_url":"https://github.com/aime-team.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# AIME MLC - Machine Learning Container management system\n\nEasily install, run and manage Docker containers for Pytorch and Tensorflow deep learning frameworks.\n\n## Features\n\n* Setup and run a specific version of Pytorch or Tensorflow with one simple command\n* Run different versions of machine learning frameworks and required libraries in parallel\n* manages required libraries (CUDA, CUDNN, CUBLAS, etc.) in containers, without compromising the host installation\n* Clear separation of user code and framework installation, test your code with a different framework version in minutes\n* multi session: open and run many shell session on a single container simultaneously\n* multi user: separate container space for each user\n* multi GPU: allocate GPUs per user, container or session\n* Runs with the same performance as a bare metal installation\n* Repository of all major deep learning framework versions as containers\n* (New): create and configure containers in interactive mode\n* (New): GPU architecture switchable by flag or an enviroment variable\n\n## Installation\n\nAIME machines come pre installed with AIME machine learning container management system for more information see: https://www.aime.info/mlc\n\nPlease read on in the [AIME MLC Installation Guide](aime-mlc-installation-guide.md) how to install AIME MLC on your PC, workstation or server.\n\n## Usage\n\nTo view detailed information about all available commands, use:\n```\nmlc -h\n```\n\nTo get help for a specific command (e.g., create), use:\n```\nmlc create -h\n```\n\n\n### Create a machine learning container\n\n**mlc create container_name framework version [-w workspace\\_dir] [-d data\\_dir] [-m models\\_dir] [-s|--script] [-arch|--architecture gpu_architecture] [-g|--num_gpus all]**\n\nCreate a new machine learning container\n\nAvailable frameworks:\n\nPytorch, Tensorflow \n\nThe following architectures are currently available: \n\nCUDA_ADA, CUDA_AMPERE, CUDA for NVIDIA GPUs and ROCM6 and ROCM5 for AMD GPUs.\n\nAvailable versions for CUDA_ADA, which corresponds to NVIDIA Ada Lovelace based GPUs (RTX 4080/4090, RTX 4500/5000/6000 Ada, L40, L40S):\n\n* Pytorch: 2.6.0, 2.5.1, 2.5.0, 2.4.0, 2.3.1-aime, 2.3.0, 2.2.2, 2.2.0, 2.1.2-aime, 2.1.1-aime, 2.1.0-aime, 2.1.0, 2.0.1-aime, 2.0.1, 2.0.0, 1.14.0a-nvidia, 1.13.1-aime, 1.13.0a-nvidia, 1.12.1-aime\n\n* Tensorflow: 2.16.1, 2.15.0, 2.14.0, 2.13.1-aime, 2.13.0, 2.12.0, 2.11.0-nvidia, 2.11.0-aime, 2.10.1-nvidia, 2.10.0-nvidia, 2.9.1-nvidia\n\nTo print the current available gpu architectures, frameworks and corresponding versions, use:\n\n```\nmlc create --info\n```\n\nExample to create a container in script mode using Pytorch 2.4.0 with the name 'my-container' and with mounted user home directory as workspace, /data and /models as data and models directory, use:\n\n```\nmlc create my-container Pytorch 2.4.0 -w /home/user_name/workspace -d /data -m /models -s -arch CUDA_AMPERE\n```\n\nTo provide greater flexibility in selecting a GPU architecture, users can specify the desired architecture for the current container using the -arch cuda_architecture flag (default: CUDA_ADA). If a fixed architecture is preferred for an entire session, it can be set by saving the desired GPU architecture in the MLC_ARCH environment variable, for example: export MLC_ARCH=CUDA_AMPERE\n\n\n### Open a machine learning container\n\n**mlc open container_name [-s|--script]**\n\nTo open the created machine learning container \"my-container\" using the script mode\n\n```\nmlc open my-container -s\n```\n\nWill output:\n\n```\n[my-container] starting container\n[my-container] opening shell to container\n\n________                               _______________\n___  __/__________________________________  ____/__  /________      __\n__  /  _  _ \\_  __ \\_  ___/  __ \\_  ___/_  /_   __  /_  __ \\_ | /| / /\n_  /   /  __/  / / /(__  )/ /_/ /  /   _  __/   _  / / /_/ /_ |/ |/ /\n/_/    \\___//_/ /_//____/ \\____//_/    /_/      /_/  \\____/____/|__/\n\n\nYou are running this container as user with ID 1000 and group 1000,\nwhich should map to the ID and group for your user on the Docker host. Great!\n\n[my-container] admin@aime01:/workspace$\n```\n\nThe container is run with the access rights of the user. To use privileged rights like for installing packages with 'apt' within the container use 'sudo'. The default is that no password is needed for sudo, to change this behaviour set a password with 'passwd'.\n\nMultiple instances of a container can be opened with mlc open. Each instance runs in its own process.\n\nTo exit an opened shell to the container type 'exit' on the command line. The last exited shell will automatically stop the container.\n\n\n### List available machine learning containers\n\n**mlc list** will list all available containers for the current user\n\n\n```\nmlc list\n```\n\nwill output for example:\n\n```\nAvailable ml-containers are:\n\nCONTAINER           FRAMEWORK                  STATUS\n[torch-vid2vid]     Pytorch-1.2.0              Up 2 days\n[tf1.15.0]          Tensorflow-1.15.0          Up 8 minutes\n[tf1-nvidia]        Tensorflow-1.14.0_nvidia   Exited (137) 1 week ago\n[tf1.13.2]          Tensorflow-1.13.2          Exited (137) 2 weeks ago\n[torch1.3]          Pytorch-1.3.0              Exited (137) 3 weeks ago\n[tf2-gpt2]          Tensorflow-2.0.0           Exited (137) 7 hours ago\n```\n\nIn case that you wish to see all containers created by all users:\n\n```\nmlc list -au\n```\n\n\n### List the stats of active machine learning containers\n\n**mlc stats** show all current running ml containers and their CPU and memory usage\n\n```\nmlc stats\n\nRunning ml-containers are:\n\nCONTAINER           CPU %               MEM USAGE / LIMIT\n[torch-vid2vid]     4.93%               8.516GiB / 63.36GiB\n[tf1.15.0]          7.26%               9.242GiB / 63.36GiB\n```\n\n### Start machine learning containers\n\n**mlc start container_name [-s|--script]** to explicitly start a container\n\n'mlc start' is a way to start the container to run installed background processes, like an installed web server, on the container without the need to open an interactive shell to it.\n\nFor opening a shell to the container just use 'mlc open', which will automatically start the container if the container is not already running.\n\n\n### Stop machine learning containers\n\n**mlc stop container_name [-s|--script] [-f|--force]** to explicitly stop a container.\n\n'mlc stop' on a container is comparable to a shutdown of a computer, all activate processes and open shells to the container will be terminated.\n\nTo force a stop on a container use:\n\n```\nmlc stop my-container -f\n```\n\nAs by the other commands, the script mode is available using the flag -s:\n\n```\nmlc stop my-container -s\n```\n\n\n### Remove/Delete a machine learning container\n\n**mlc remove container_name [-s|--script] [-f|--force]** to remove the container.\n\nWarning: the container will be unrecoverable deleted only data stored in the /workspace directory will be kept. Only use to clean up containers which are not needed any more.\n\n```\nmlc remove my-container\n```\n\n### Update MLC\n\n**mlc update-sys** to update the container managment system to latest version.\n\nThe container system and container repo will be updated to latest version. Run this command to check if new framework versions are available. On most systems privileged access (sudo password) is required to do so.\n\n```\nmlc update-sys\n```\n\nThe force option (-f) is available too.\n\n## Supported ML containers\n\n### Pytorch Containers\n\n| Container Name | GPU Arch. | Build  | Pytorch  | Ubuntu Version | Python Version | Package Manager | CUDA Version | CuDNN Version | NVIDIA driver version |\n|:----------------:|:--------:|:--------:|:----------:|:----------------:|:----------------:|:-----------------:|:--------------:|:---------------:|:-----------------------:|\n| 2.6.0 | CUDA_ADA | AIME | 2.6.0 | 22.04 | 3.10.12 | pip 22.0.2 | 12.4.131 | 9.1.0.70 | 550.54.15 |\n| 2.6.0 | CUDA_AMPERE | AIME | 2.6.0 | 22.04 | 3.10.12 | pip 22.0.2 | 11.8.89 | 8.9.6.50 | 520.61.05 |\n| 2.5.1 | CUDA_ADA | AIME | 2.5.1 | 22.04 | 3.10.12 | pip 22.0.2 | 12.1.105 | 8.9.0.131 | 530.30.02 |\n| 2.5.1 | CUDA_AMPERE | AIME | 2.5.1 | 22.04 | 3.10.12 | pip 22.0.2 | 11.8.89 | 8.9.6.50 | 520.61.05 |\n| 2.5.0 | CUDA_ADA | Official | 2.5.0 | 22.04 | 3.11.10 | conda 24.9.2 | 12.1.105 | 9.1.0 | 530.30.02 |\n| 2.5.0 | CUDA_AMPERE | Official | 2.5.0 | 22.04 | 3.11.10 | conda 24.9.2 | 11.8.89 | 9.1.0 | 520.61.05 |\n| 2.4.0 | CUDA_ADA | Official | 2.4.0 | 22.04 | 3.11.9 | conda 24.5 | 12.1.105 | 9.1.0 | 530.30.02 |\n| 2.4.0 | CUDA_AMPERE | Official | 2.4.0 | 22.04 | 3.11.9 | conda 24.5 | 11.8.89 | 9.1.0 | 520.61.05 |\n| 2.3.1-aime | CUDA_ADA | AIME | 2.3.1 | 22.04 | 3.10.12 | pip 22.0.2 | 12.1.105 | 8.9.7.29 | 530.30.02 |\n| 2.3.1-aime | CUDA_AMPERE | AIME | 2.3.1 | 22.04 | 3.10.12 | pip 22.0.2 | 11.8.87 | 8.9.6.50 | 520.61.05 |\n| 2.3.0 | CUDA_ADA | Official | 2.3.0 | 22.04 | 3.10.14 | conda 23.5.2 | 12.1.105 | 8.9.7.29 | 530.30.02 |\n| 2.2.2 | CUDA_ADA | AIME | 2.2.2 | 22.04 | 3.10.12 | pip 22.0.2 | 12.1.105 | 8.9.7.29 | 530.30.02 |\n| 2.2.0 | CUDA_ADA | Official | 2.2.0 | 22.04 | 3.10.13 | conda 23.9.0 | 12.1.105 | 8.9.0.131 | 530.30.02 |\n| 2.1.2-aime | CUDA_ADA | AIME | 2.1.2 | 22.04 | 3.10.12 | pip 22.0.2 | 12.1.105 | 8.9.0.131 | 530.30.02 |\n| 2.1.1-aime | CUDA_ADA | AIME | 2.1.1 | 22.04 | 3.10.12 | pip 22.0.2 | 12.1.105 | 8.9.0.131 | 530.30.02 |\n| 2.1.0-aime | CUDA_ADA, CUDA_AMPERE | AIME | 2.1.0 | 22.04 | 3.10.12 | pip 22.0.2 | 11.8.89 | 8.9.0.131 | 520.61.05 |\n| 2.1.0 | CUDA_ADA | Official | 2.1.0 | 22.04 | 3.10.13 | conda 23.9.0 | 12.1.105 | 8.9.0.131 | 530.30.02 |\n| 2.0.1-aime | CUDA_ADA, CUDA_AMPERE | AIME | 2.0.1 | 22.04 | 3.10.12 | pip 22.0.2 | 11.8.89 | 8.9.0.131 | 520.61.05 |\n| 2.0.1 | CUDA_ADA, CUDA_AMPERE | Official | 2.0.1 | 20.04          | 3.8.10         | pip 20.0.2      | 11.8.89      | 8.9.0.131     | 520.61.05             |\n| 2.0.0 | CUDA_ADA, CUDA_AMPERE | Official | 2.0.0 | 20.04          | 3.8.10         | pip 20.0.2      | 11.8.89      | 8.6.0.163     | 520.61.05             |\n| 1.14.0a-nvidia | CUDA_ADA | NVIDIA   | 1.14.0a0          | 20.04          | 3.8.10         | pip 21.2.4      | 12.0.146     | 8.7.0.84      | 525.85.11\n| 1.13.1-aime    | CUDA_ADA | AIME     | 1.13.1            | 20.04          | 3.8.10         | pip 20.0.2      | 11.8.89      | 8.6.0.163     | 520.61.05\n| 1.13.0a-nvidia | CUDA_ADA | NVIDIA   | 1.13.0a0          | 20.04          | 3.8.13         | pip 21.2.4      | 11.8.89      | 8.6.0.163     | 520.61.03\n| 1.12.1-aime    | CUDA_ADA | AIME     | 1.12.1            | 20.04          | 3.8.10         | pip 20.0.2      | 11.8.89      | 8.6.0.163     | 520.61.05           \n\n### Tensorflow Containers\n\n| Container Name | GPU Arch. | Build  | Tensorflow Version | Ubuntu Version | Python Version | Package Manager | CUDA Version | CuDNN Version | NVIDIA driver version |\n|:----------------:|:--------:|:--------:|:--------------------:|:----------------:|:----------------:|:-----------------:|:--------------:|:---------------:|:-----------------------:|\n| 2.16.1 | CUDA_ADA | Official | 2.16.1 | 22.04 | 3.11.0rc1 | pip 24.0 | 12.3.107 | 8.9.6.50 | 545.23.06 |\n| 2.15.0 | CUDA_ADA | Official | 2.15.0 | 22.04 | 3.11.0rc1 | pip 23.3.1 | 12.3.103 | 8.9.6.50 | 545.23.06 |\n| 2.14.0 | CUDA_ADA, CUDA_AMPERE | Official | 2.14.0 | 22.04 | 3.11.0rc1 | pip 23.2.1 | 11.8.89 | 8.6.0.163 | 525.85.12 |\n| 2.13.1-aime | CUDA_ADA, CUDA_AMPERE | AIME | 2.13.1 | 22.04 | 3.10.12 | pip 22.0.2 | 11.8.89 | 8.9.0.131 | 520.61.05 |\n| 2.13.0 | CUDA_ADA, CUDA_AMPERE | Official | 2.13.0 | 20.04 | 3.8.10 | pip 23.0.1 | 11.8.89 | 8.6.0.163 | 525.85.12 |\n| 2.12.0 | CUDA_ADA, CUDA_AMPERE | Official | 2.12.0 | 20.04 | 3.8.10 | pip 23.0.1 | 11.8.89 | 8.6.0.163 | 525.85.12 |\n| 2.11.0-nvidia  | CUDA_ADA | NVIDIA | 2.11.0             | 20.04          | 3.8.10         | pip 22.3.1      | 12.0.146     | 8.7.0.84      | 525.85.12 |\n| 2.11.0-aime    | CUDA_ADA | AIME   | 2.11.0             | 20.04          | 3.8.10         | pip 20.0.2      | 11.8.89      | 8.6.0.163     | 520.61.05 |\n| 2.10.1-nvidia  | CUDA_ADA | NVIDIA | 2.10.1             | 20.04          | 3.8.10         | pip 22.3.1      | 11.8.89      | 8.7.0.84      | 520.61.05             |\n| 2.10.0-nvidia  | CUDA_ADA | NVIDIA | 2.10.0             | 20.04          | 3.8.10         | pip 22.2.2      | 11.8.89      | 8.6.0.163     | 520.61.05             |\n| 2.9.1-nvidia   | CUDA_ADA | NVIDIA | 2.9.1              | 20.04          | 3.8.10         | pip 22.2.2      | 11.8.89      | 8.6.0.163     | 520.61.03             |\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faime-team%2Faime-mlc","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Faime-team%2Faime-mlc","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faime-team%2Faime-mlc/lists"}