{"id":15157721,"url":"https://github.com/liuyuweitarek/downgrade-upgrade-colab-python","last_synced_at":"2026-01-21T16:33:29.280Z","repository":{"id":254744270,"uuid":"847379348","full_name":"liuyuweitarek/downgrade-upgrade-colab-python","owner":"liuyuweitarek","description":"Creating a Specific Version of Python and PyTorch Environment on Colab.","archived":false,"fork":false,"pushed_at":"2024-08-31T10:53:43.000Z","size":45,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-02-13T16:53:50.830Z","etag":null,"topics":["customized-project","google-colab","python","pytorch","tensorflow"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/liuyuweitarek.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","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":"2024-08-25T16:55:20.000Z","updated_at":"2024-08-31T10:53:45.000Z","dependencies_parsed_at":"2024-08-25T20:23:07.939Z","dependency_job_id":"3574b556-ac74-45ae-91c8-e12b802187fb","html_url":"https://github.com/liuyuweitarek/downgrade-upgrade-colab-python","commit_stats":{"total_commits":10,"total_committers":1,"mean_commits":10.0,"dds":0.0,"last_synced_commit":"24438645759b4ac47f9b23552193a1606918bfb1"},"previous_names":["liuyuweitarek/downgrade-upgrade-colab-python"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liuyuweitarek%2Fdowngrade-upgrade-colab-python","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liuyuweitarek%2Fdowngrade-upgrade-colab-python/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liuyuweitarek%2Fdowngrade-upgrade-colab-python/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/liuyuweitarek%2Fdowngrade-upgrade-colab-python/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/liuyuweitarek","download_url":"https://codeload.github.com/liuyuweitarek/downgrade-upgrade-colab-python/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247668451,"owners_count":20976225,"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":["customized-project","google-colab","python","pytorch","tensorflow"],"created_at":"2024-09-26T20:01:59.502Z","updated_at":"2026-01-21T16:33:24.254Z","avatar_url":"https://github.com/liuyuweitarek.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Creating a Specific Version of Python and PyTorch Environment on Colab\n\nThis document will guide free users in creating a specific version of Python, PyTorch and Tensorflow execution environment on Google Colab. \n\nBased on the steps provided here, you can create any Python(version \u003e= 3.7) environment in Google Colab. \n\n**Noted that this method has limitations.** \n\nWith this method, cell cannot directly execute Python code from the environment you create. You need to activate the created environment at the start of each cell execution, which is equivalent to running your command in the terminal.\n\ni.e. Cell will be like below to execute your code:\n\n```\n%%shell\neval \"$(conda shell.bash hook)\"\nconda activate myenv\npython main.py\n```\n\n## Prerequisite\n\n### 1. Check CUDA versions\n  Please refer to [this table](https://www.tensorflow.org/install/source?hl=zh-tw#gpu) to find the Python Version you're using and the corresponding supported CUDA Version. \n\nP.S. No matter you are using PyTorch or TensorFlow.\n \n### 2. Check your package version \n\n#### For Pytorch\n\n  Check the table below, and look for whether the package is in this [package source list](https://download.pytorch.org/whl/torch/).\n\n  For example:\n  - Python 3.7(cp37)\n  - PyTorch-1.7.1\n  - CUDA Version 10.1(cu101)\n  - and google colab is linux based system(linux_x86_64).\n  \n  I then should found `torch-1.7.1+cu101-cp37-cp37m-linux_x86_64.whl` in the list, which cound be installed in command like:\n    \n  ```bash\n  $ python -m pip install torch==1.7.1+cu101 --extra-index-url https://download.pytorch.org/whl --no-cache-dir\n  ```\n\n\u003ctable\u003e\n  \u003cthead\u003e\n    \u003ctr\u003e\n      \u003cth\u003eTorch Version\u003c/th\u003e\n      \u003cth\u003eAvailable CUDA Version\u003c/th\u003e\n      \u003cth\u003ePython Version\u003c/th\u003e\n    \u003c/tr\u003e\n  \u003c/thead\u003e\n  \u003ctbody\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e2.0.1\u003c/td\u003e\n      \u003ctd\u003ecu117, cu118\u003c/td\u003e\n      \u003ctd\u003ecp38, cp39, cp310, cp311\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e2.0.0\u003c/td\u003e\n      \u003ctd\u003ecu117, cu118\u003c/td\u003e\n      \u003ctd\u003ecp38, cp39, cp310, cp311\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.13.1\u003c/td\u003e\n      \u003ctd\u003ecu116, cu117\u003c/td\u003e\n      \u003ctd\u003ecp37, cp38, cp39, cp310\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.13.0\u003c/td\u003e\n      \u003ctd\u003ecu116, cu117\u003c/td\u003e\n      \u003ctd\u003ecp37, cp38, cp39, cp310\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.12.1\u003c/td\u003e\n      \u003ctd\u003ecu113, cu116\u003c/td\u003e\n      \u003ctd\u003ecp37, cp38, cp39, cp310\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.12.0\u003c/td\u003e\n      \u003ctd\u003ecu113, cu116\u003c/td\u003e\n      \u003ctd\u003ecp37, cp38, cp39, cp310\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.11.0\u003c/td\u003e\n      \u003ctd\u003ecu113, cu115\u003c/td\u003e\n      \u003ctd\u003ecp37, cp38, cp39, cp310\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.10.2\u003c/td\u003e\n      \u003ctd\u003ecu102, cu111, cu113\u003c/td\u003e\n      \u003ctd\u003ecp36, cp37, cp38, cp39\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.10.1\u003c/td\u003e\n      \u003ctd\u003ecu102, cu111, cu113\u003c/td\u003e\n      \u003ctd\u003ecp36, cp37, cp38, cp39\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.10.0\u003c/td\u003e\n      \u003ctd\u003ecu102, cu111, cu113\u003c/td\u003e\n      \u003ctd\u003ecp36, cp37, cp38, cp39\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.9.1\u003c/td\u003e\n      \u003ctd\u003ecu102, cu111\u003c/td\u003e\n      \u003ctd\u003ecp36, cp37, cp38, cp39\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.9.0\u003c/td\u003e\n      \u003ctd\u003ecu102, cu111\u003c/td\u003e\n      \u003ctd\u003ecp36, cp37, cp38, cp39\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.8.1\u003c/td\u003e\n      \u003ctd\u003ecu101, cu102, cu111\u003c/td\u003e\n      \u003ctd\u003ecp36, cp37, cp38, cp39\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.8.0\u003c/td\u003e\n      \u003ctd\u003ecu101, cu111\u003c/td\u003e\n      \u003ctd\u003ecp36, cp37, cp38, cp39\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.7.1\u003c/td\u003e\n      \u003ctd\u003ecu101, cu110\u003c/td\u003e\n      \u003ctd\u003ecp36, cp37, cp38, cp39\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.7.0\u003c/td\u003e\n      \u003ctd\u003ecu101, cu110\u003c/td\u003e\n      \u003ctd\u003ecp36, cp37, cp38\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.6.0\u003c/td\u003e\n      \u003ctd\u003ecu101\u003c/td\u003e\n      \u003ctd\u003ecp36, cp37, cp38\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.5.1\u003c/td\u003e\n      \u003ctd\u003ecu92, cu101\u003c/td\u003e\n      \u003ctd\u003ecp35, cp36, cp37, cp38\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.5.0\u003c/td\u003e\n      \u003ctd\u003ecu92, cu101\u003c/td\u003e\n      \u003ctd\u003ecp35, cp36, cp37, cp38\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.4.0\u003c/td\u003e\n      \u003ctd\u003ecu92\u003c/td\u003e\n      \u003ctd\u003ecp35, cp36, cp37, cp38\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.3.1\u003c/td\u003e\n      \u003ctd\u003ecu92\u003c/td\u003e\n      \u003ctd\u003ecp35, cp36, cp37\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.3.0\u003c/td\u003e\n      \u003ctd\u003ecu92\u003c/td\u003e\n      \u003ctd\u003ecp35, cp36, cp37\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e1.2.0\u003c/td\u003e\n      \u003ctd\u003ecu92\u003c/td\u003e\n      \u003ctd\u003ecp35, cp36, cp37\u003c/td\u003e\n    \u003c/tr\u003e\n  \u003c/tbody\u003e\n\u003c/table\u003e\n\n#### For Tensorflow\n\nCheck out [the version table](https://www.tensorflow.org/install/source?hl=zh-tw#gpu). \n\nFor example,\n- Python 3.7\n- Tensorflow 2.1.0\n\nAt `requirements.txt`,\n\n```\ntensorflow-gpu==2.1.0\nprotobuf==3.20.1\n```\n\n### 3. Install cudnn + CUDA\n\nBased on the `CUDA Version` you determined above, find the cudnn package to install:\n\nFor example, I chose CUDA 10.1 above, so after checkouting the table below, I download `cudnn-10.1-linux-x64-v7.6.5.32.tgz`.\n\nI chose `v7.6.5.32` because many references use this version without any issues. However, in principle, you just need to select one that matches the version compatibility listed below.\n\nBelow are the download links for CUDA versions \u003c= 11. \n\n\u003e _If you are using CUDA version \u003e= 12, you might consider downloading it directly from the official website._\n\n|Cudnn Version|Sources|\n| --- | --- |\n|v5 - v8.8.0| https://developer.download.nvidia.com/compute/redist/cudnn |\n|v8.4.0 - v9.3.0| https://developer.download.nvidia.com/compute/cudnn/redist/cudnn/linux-x86_64 |\n\n\u003ctable\u003e\u003cthead\u003e\u003ctr\u003e\u003cth\u003ecuDNN\u003c/th\u003e\u003cth\u003eCUDA\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.4.0 (April 1st, 2022)\u003c/td\u003e\u003ctd\u003eCUDA 11.x\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.4.0 (April 1st, 2022)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.3.3 (March 18th, 2022)\u003c/td\u003e\u003ctd\u003eCUDA 11.5\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.3.3 (March 18th, 2022)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.3.2 (January 10th, 2022)\u003c/td\u003e\u003ctd\u003eCUDA 11.5\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.3.2 (January 10th, 2022)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.3.1 (November 22nd, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 11.5\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.3.1 (November 22nd, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.3.0 (November 3rd, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 11.5\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.3.0 (November 3rd, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.2.4 (September 2nd, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 11.4\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.2.4 (September 2nd, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.2.2 (July 6th, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 11.4\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.2.2 (July 6th, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.2.1 (June 7th, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 11.x\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.2.1 (June 7th, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.2.0 (April 23rd, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 11.x\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.2.0 (April 23rd, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.1.1 (Feburary 26th, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 11.0,11.1 and 11.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.1.1 (Feburary 26th, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.1.0 (January 26th, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 11.0,11.1 and 11.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.1.0 (January 26th, 2021)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.5 (November 9th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 11.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.5 (November 9th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 11.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.5 (November 9th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.5 (November 9th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 10.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.4 (September 28th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 11.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.4 (September 28th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 11.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.4 (September 28th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.4 (September 28th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 10.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.3 (August 26th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 11.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.3 (August 26th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.3 (August 26th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 10.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.2 (July 24th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 11.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.2 (July 24th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.2 (July 24th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 10.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.1 RC2 (June 26th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 11.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v8.0.1 RC2 (June 26th, 2020)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.5 (November 18th, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.5 (November 5th, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.5 (November 5th, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.5 (November 5th, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.5 (November 5th, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.4 (September 27, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.4 (September 27, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.4 (September 27, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.4 (September 27, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.3 (August 23, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.3 (August 23, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.3 (August 23, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.3 (August 23, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.2 (July 22, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.2 (July 22, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.2 (July 22, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.2 (July 22, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.1 (June 24, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.1 (June 24, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.1 (June 24, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.1 (June 24, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.0 (May 20, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.0 (May 20, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.0 (May 20, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.6.0 (May 20, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.5.1 (April 22, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.5.1 (April 22, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.5.1 (April 22, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.5.1 (April 22, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.5.0 (Feb 25, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.5.0 (Feb 21, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 10.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.5.0 (Feb 21, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.5.0 (Feb 21, 2019)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.4.2 (Dec 14, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 10.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.4.2 (Dec 14, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.4.2 (Dec 14, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.4.1 (Nov 8, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 10.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.4.1 (Nov 8, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.4.1 (Nov 8, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.3.1 (Sept 28, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 10.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.3.1 (Sept 28, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.3.1 (Sept 28, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.3.0 (Sept 19, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 10.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.3.0 (Sept 19, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.2.1 (August 7, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.1.4 (May 16, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.1.4 (May 16, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.1.4 (May 16, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 8.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.1.3 (April 17, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.1.3 (April 17, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.1.3 (April 17, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 8.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.1.2 (Mar 21, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.1 \u0026amp; 9.2\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.1.2 (Mar 21, 2018)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.0.5 (Dec 11, 2017)\u003c/td\u003e\u003ctd\u003eCUDA 9.1\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.0.5 (Dec 5, 2017)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.0.5 (Dec 5, 2017)\u003c/td\u003e\u003ctd\u003eCUDA 8.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v7.0.4 (Nov 13, 2017)\u003c/td\u003e\u003ctd\u003eCUDA 9.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v6.0 (April 27, 2017)\u003c/td\u003e\u003ctd\u003eCUDA 8.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v6.0 (April 27, 2017)\u003c/td\u003e\u003ctd\u003eCUDA 7.5\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v5.1 (Jan 20, 2017)\u003c/td\u003e\u003ctd\u003eCUDA 8.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v5.1 (Jan 20, 2017)\u003c/td\u003e\u003ctd\u003eCUDA 7.5\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v5 (May 27, 2016)\u003c/td\u003e\u003ctd\u003eCUDA 8.0\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v5 (May 12, 2016)\u003c/td\u003e\u003ctd\u003eCUDA 7.5\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v4 (Feb 10, 2016)\u003c/td\u003e\u003ctd\u003eCUDA 7.0 and later.\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v3 (September 8, 2015)\u003c/td\u003e\u003ctd\u003eCUDA 7.0 and later.\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v2 (March 17,2015)\u003c/td\u003e\u003ctd\u003eCUDA 6.5 and later.\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd\u003ecuDNN v1 (cuDNN 6.5 R1)\u003c/td\u003e\u003ctd\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\n\n## Get started\n\nAs proof as concept, I Use Miniconda create a Python 3.7 environment, which can use both PyTorch 1.7.1 and tensorflow 2.1.0 version here.\n\n1. Clone the project\n\n    ```bash\n    git clone https://github.com/liuyuweitarek/downgrade-upgrade-colab-python.git custom_env_colab\n    ```\nIf have a better approach, feel free to share it with me or submit a PR. Thank you!\n\n2. Install `cudnn-10.1-linux-x64-v7.6.5.32.tgz` from [NVIDIA website](https://developer.download.nvidia.com/compute/redist/cudnn/v7.6.5/cudnn-10.1-linux-x64-v7.6.5.32.tgz). Here is the [backup](https://drive.google.com/file/d/1R54KoKk16CcpFu3tR80dEyC_aB8UOyXc/view?usp=sharing). Place in the same folder as `custom_env_colab`.\nFurther,\n\n3. Place the project in your Google Drive.\n\n4. Follow the instructions in the `Creating Specific Version of Python and PyTorch Environment on Colab.ipynb` notebook.\n\n## Q\u0026A and Notes\n1. How to deal with Colab time limit?\n\n    This document does not overcome the time limit issue for free users, who will need to wait for a specific period before using it again. Therefore, please **make good use of Checkpoint to save and continue training progress**.\n\n2. Please use the version of Miniconda installer which is higher than the Python version you wish to use to create the virtual environment.\nFurthermore,\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fliuyuweitarek%2Fdowngrade-upgrade-colab-python","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fliuyuweitarek%2Fdowngrade-upgrade-colab-python","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fliuyuweitarek%2Fdowngrade-upgrade-colab-python/lists"}