{"id":23636678,"url":"https://github.com/0xscratch/aws-deepracer","last_synced_at":"2025-07-02T10:33:01.917Z","repository":{"id":174881720,"uuid":"525348688","full_name":"0xScratch/Aws-Deepracer","owner":"0xScratch","description":null,"archived":false,"fork":false,"pushed_at":"2022-08-16T11:27:00.000Z","size":82622,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-06-12T11:55:33.918Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit-0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/0xScratch.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}},"created_at":"2022-08-16T11:24:15.000Z","updated_at":"2022-08-16T11:29:08.000Z","dependencies_parsed_at":null,"dependency_job_id":"ade19bd5-2107-495a-a8c9-dabf15673b91","html_url":"https://github.com/0xScratch/Aws-Deepracer","commit_stats":null,"previous_names":["aryan9592/aws-deepracer","0xscratch/aws-deepracer"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/0xScratch/Aws-Deepracer","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0xScratch%2FAws-Deepracer","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0xScratch%2FAws-Deepracer/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0xScratch%2FAws-Deepracer/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0xScratch%2FAws-Deepracer/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/0xScratch","download_url":"https://codeload.github.com/0xScratch/Aws-Deepracer/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0xScratch%2FAws-Deepracer/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":263121098,"owners_count":23416934,"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":[],"created_at":"2024-12-28T06:14:19.080Z","updated_at":"2025-07-02T10:33:01.894Z","avatar_url":"https://github.com/0xScratch.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# deepracer-analysis\n\nThis is a set of notebooks and utilities to enable analysis of logs for AWS DeepRacer.\n\nThis project is a redo of analysis solutions provided in the\n[AWS DeepRacer Workshops repository](https://github.com/aws-samples/aws-deepracer-workshops).\n\nThere are a few motivations leading a decision to reorganise the repository\n* Having a community version of the project in a folder on a branch of a fork\nof the original git repository has been causing issues when looking for it\n* Jupyter notebooks are difficult to manage through source control as a generated file in\njson format is hardly readable at all. This needed a new approach\n* The project relied on a bunch of Python files which acted as a bag for code. There was\na need to extract it into a separate project to maintain the versioning and apply some\nhopefully good practices\n\nSeparate repository makes it easier to find it. Jupyter combined with\n[Jupytext](https://github.com/mwouts/jupytext) enables maintaining the notebook as\na set of Python files from which a notebook can then be genereted.\nFinally, the project files have been moved to\n[DeepRacer utils](https://github.com/aws-deepracer-community/deepracer-utils).\n\n## Required setup\n\nIt is assumed that you have aws-cli installed and configured in your account.\nYou will find instructions in [documentation](https://docs.aws.amazon.com/cli/latest/userguide/cli-chap-configure.html).\nWithout this the notebooks will not fail to load but fetching logs from AWS\nwill not work. If you realise this after running the notebook, you can open\na terminal withing Jupyter Lab and perform the configuration.\n\nIf this happens in Docker, just bear in mind that ~/.aws folder will be created\nfor you with root privileges - this may lead to permissions problems at some point.\n\n## Using the notebooks with Docker\n\nThe recommended way to work with this project is by using Docker containers. Containers\nprovide an isolated, disposable environment for your use. If you however prefer not to use\nDocker (or are using Windows), see \"Using the notebooks without Docker\" below.\n\nSince you're using DeepRacer Analysis, chances are you've already got Docker installed.\nIf not, find instructions in [Docker documentation](https://docs.docker.com/install/).\n\nDocker setup comes with Jupytext configured.\n\n### Building the Docker image\n\nBefore you run your notebooks, you will have to build the docker image:\n```\nbin/build-docker-image.sh\n```\nI'd recommend that you do it every time when you pull changes from the git repository.\n\nThis builds a Docker image on top of a jupyter-minimal image and installs required dependencies.\n\n### Starting the analysis\n\nTo start using the analysis you have to first start the container and then open the notebook\nin a browser. The startup script starts Jupyter Notebook but is you add `lab` argument\nit will open Jupyter Lab - this is my preferred way\n```\nbin/start.sh lab\nbin/open-notebook.sh\n```\nIf you're running on a remote system, you can use `url-to-notebook.sh` to obtain a url with\na token to open in your browser. You can provide your url as an argument, otherwise you will\nget a localhost address:\n```\nbin/url-to-notebook.sh http://someurl.com:8888\n```\nwill return\n```\nhttp://someulr.com:8888/?token=123fab41...\n```\nif the container is running.\n\n## Using the notebooks without Docker\n\nThe notebooks require Jupyter to run, together with deepracer-utils. While not needed\nfor using the notebooks, it's worth to also have Jupytext installed.\n\nIf you only plan to use the notebooks, I recommend that you make a copy of them to enable\nseamless pulls of any updates.\n\nIf you pull latest changes for the notebooks, do also run\n```\npip install --upgrade -r requirements.txt\n```\nin your venv. This way you will also get upgrades on the requirements.\n\n### Running\n```\npython3 -m venv venv\nsource venv/bin/activate\npip install --upgrade -r requirements.txt\njupyter lab\n```\n\n### Modifying the notebooks\nIf you want to use the notebooks as a user and don't intend to submit changes,\nsimply use them through Jupyter Notebook or Jupyter Lab.\n\nIf you would like to submit changes to a notebook however, follow instruction in the\n[Jupytext README](https://github.com/mwouts/jupytext) to enable pairing of the notebook\nwith a light script. This means that any changes you apply to the notebook or the .py\nfile paired with it will be synched.\n\nWhen applying changes to the notebook, make sure you can use them with the sample log\nresources and at the end of work restart the Kernel and run all the cells to provide\na clean view in the notebook.\n\n## Roadmap\n* [x] Recreate the training and evaluation notebooks on top of the deepracer-utils\n* [x] Apply changes from the original notebook commited since the creation of the log analysis branch\n* [x] Add docker runtime\n* [ ] Redo log analysis challenge PRs and apply them to notebooks\n* [ ] Prepare simpler, specialised notebooks for everyday use\n* [ ] Prepare a tutorial on how to use and contribute to a notebook\n\n## Credits\nWe would like to thank:\n* AWS employees who have given birth to these tools as part of the\n[AWS DeepRacer Workshops repository](https://github.com/aws-samples/aws-deepracer-workshops).\n* All involved AWS DeepRacer Community members to contributed to its development\n\n## License\nThis project retains the license of the \n[aws-deepracer-workshops](https://github.com/aws-samples/aws-deepracer-workshops)\nproject which has been forked for the initial Community contributions.\nOur understanding is that it is a license more permissive than the MIT license\nand allows for removing of the copyright headers.\n\nUnless clearly sated otherwise, this license applies to all files in this repository.\n\n## Troubleshooting\n\nIf you face problems, do reach out to the [AWS DeepRacer Community](http://join.deepracing.io).\nChannel #dr-training-log-analysis has been created for this purpose.\nWhen you face an issue, it is worth running `pip freeze` and saving the output as it may be\ndue to a specific version of the dependencies installed.\n\n## Contact\nYou can contact Tomasz Ptak through the Community Slack: http://join.deepracing.io.\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F0xscratch%2Faws-deepracer","html_url":"https://awesome.ecosyste.ms/projects/github.com%2F0xscratch%2Faws-deepracer","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F0xscratch%2Faws-deepracer/lists"}