{ "id": 324554, "title": "Ideas for Generating New Data 💡 ", "url": "/competitions/AI4Code/discussion/324554", "author_name": "", "post_date": "2022-05-12T08:57:08.415318500Z", "votes": 50, "comment_count": 18, "views": 0, "content": "

New data can be generated for training using the following approaches:

\n

1) Code -> Comments 📚

\n

\"\"
\nWe can use Docly to automatically generate the comments for our Python code.

\n

2) Comments -> Code 🚀

\n

\"\"
\nGithub Copilot can aid us in generating code for the code description we provide.

\n

🌟 Bonus: External data (?)

\n

Additionally, there was a challenge hosted on Github a few years ago called the CodeSearchNet challenge which was based on Semantic Code Search.

\n

\"\"

\n

The fully preprocessed Corpus for Python code is available for download on Amazon S3. It contains about 6 million functions.

\n

Feel free to share any ideas that you might have 🔮

", "messages": [ { "id": "1785608", "postDate": "05/12/2022 08:57:08", "content": "

New data can be generated for training using the following approaches:

\n

1) Code -> Comments 📚

\n

\"\"
\nWe can use Docly to automatically generate the comments for our Python code.

\n

2) Comments -> Code 🚀

\n

\"\"
\nGithub Copilot can aid us in generating code for the code description we provide.

\n

🌟 Bonus: External data (?)

\n

Additionally, there was a challenge hosted on Github a few years ago called the CodeSearchNet challenge which was based on Semantic Code Search.

\n

\"\"

\n

The fully preprocessed Corpus for Python code is available for download on Amazon S3. It contains about 6 million functions.

\n

Feel free to share any ideas that you might have 🔮

", "rawMarkdown": "New data can be generated for training using the following approaches:\n## 1) Code -> Comments 📚\n![](https://i.imgur.com/TgyH4vb.png)\nWe can use [Docly](http://thedocly.io/) to automatically generate the comments for our Python code.\n\n## 2) Comments -> Code 🚀\n![](https://i.imgur.com/ivYQLyZ.jpg)\nGithub Copilot can aid us in generating code for the code description we provide.\n\n## 🌟 Bonus: External data (?)\nAdditionally, there was a challenge hosted on Github a few years ago called the [CodeSearchNet challenge](https://github.blog/2019-09-26-introducing-the-codesearchnet-challenge/) which was based on Semantic Code Search.\n\n![](https://i.imgur.com/mor3WpH.png)\n\nThe [fully preprocessed Corpus for Python code](https://s3.amazonaws.com/code-search-net/CodeSearchNet/v2/python.zip) is available for download on Amazon S3. It contains about **6 million functions**. \n\nFeel free to share any ideas that you might have 🔮", "votes": null }, { "id": "1786035", "postDate": "05/12/2022 15:12:34", "content": "

@ruchi798 Thank you.
\nCould you please explain more about the 1st option?

", "rawMarkdown": "ruchi798 Thank you.\nCould you please explain more about the 1st option?", "votes": null }, { "id": "1786087", "postDate": "05/12/2022 15:42:26", "content": "

Hellooo!

\n

Docly automatically generates comments for the code blocks that we write.

\n

It can be installed using:
\npip install docly

", "rawMarkdown": "Hellooo!\n\nDocly automatically generates comments for the code blocks that we write.\n\nIt can be installed using: \n`pip install docly`", "votes": null }, { "id": "1786539", "postDate": "05/13/2022 02:08:05", "content": "

Just for curiosity, is data generated from copilot safe from license problems?

", "rawMarkdown": "Just for curiosity, is data generated from copilot safe from license problems?", "votes": null }, { "id": "1786616", "postDate": "05/13/2022 04:15:28", "content": "

Docly is awesome. Thanks for sharing @ruchi798

", "rawMarkdown": "**Docly** is awesome. Thanks for sharing @ruchi798", "votes": null }, { "id": "1786998", "postDate": "05/13/2022 13:58:36", "content": "

Thanks for sharing!! Just wondering, is it inside the competition rules to use external data?

", "rawMarkdown": "Thanks for sharing!! Just wondering, is it inside the competition rules to use external data?", "votes": null }, { "id": "1789580", "postDate": "05/14/2022 03:06:37", "content": "

Hello, @ruchi798 thanks for sharing; this is super useful, even outside of this competition.

", "rawMarkdown": "Hello, @ruchi798 thanks for sharing; this is super useful, even outside of this competition.", "votes": null }, { "id": "1790361", "postDate": "05/14/2022 20:17:56", "content": "

Is Dockly generating DocString you mean?

", "rawMarkdown": "Is **Dockly** generating DocString you mean?", "votes": null }, { "id": "1791543", "postDate": "05/16/2022 05:52:02", "content": "

I was wondering how we can use Docly in generating data ….

", "rawMarkdown": "I was wondering how we can use Docly in generating data ....", "votes": null }, { "id": "1792010", "postDate": "05/16/2022 14:55:05", "content": "

good work, thanks for sharing

", "rawMarkdown": "good work, thanks for sharing", "votes": null }, { "id": "1792180", "postDate": "05/16/2022 18:30:40", "content": "

Hi @ruchi798 Thanks for the mentions. I created Docly three years ago. It was a very early stage product I am happy to see people still remembers it :) I am also thinking to take part in this competition.

", "rawMarkdown": "Hi @ruchi798 Thanks for the mentions. I created Docly three years ago. It was a very early stage product I am happy to see people still remembers it :) I am also thinking to take part in this competition.", "votes": null }, { "id": "1792207", "postDate": "05/16/2022 18:46:57", "content": "

So nice to hear from you @rcshubhadeep 🎉
\nAll the best for the competition!

", "rawMarkdown": "So nice to hear from you @rcshubhadeep 🎉\nAll the best for the competition!", "votes": null }, { "id": "1793476", "postDate": "05/18/2022 00:07:33", "content": "

Thank you for sharing

", "rawMarkdown": "Thank you for sharing", "votes": null }, { "id": "1794138", "postDate": "05/18/2022 14:24:48", "content": "

Another awesome idea:

\n

https://thisaidoesnotexist.com/

", "rawMarkdown": "Another awesome idea:\n\nhttps://thisaidoesnotexist.com/", "votes": null }, { "id": "1794307", "postDate": "05/18/2022 16:54:26", "content": "

Useful tools, I was knowing none of them.
\nThanks for sharing

", "rawMarkdown": "Useful tools, I was knowing none of them.\nThanks for sharing", "votes": null }, { "id": "1794375", "postDate": "05/18/2022 18:25:59", "content": "

Good work for that job

", "rawMarkdown": "Good work for that job", "votes": null }, { "id": "1794749", "postDate": "05/19/2022 06:07:05", "content": "

Thanks for sharing !

", "rawMarkdown": "Thanks for sharing !", "votes": null }, { "id": "1800229", "postDate": "05/24/2022 19:19:27", "content": "

good work, thanks for sharing

", "rawMarkdown": "good work, thanks for sharing", "votes": null }, { "id": "1804639", "postDate": "05/29/2022 09:52:16", "content": "

Thank you for sharing, Is GPT models work better in these type use cases?

", "rawMarkdown": "Thank you for sharing, Is GPT models work better in these type use cases?", "votes": null } ], "comments": [ { "id": 1786035, "author_name": "samanemami", "author_url": "", "post_date": "05/12/2022 15:12:34", "content": "

@ruchi798 Thank you.
\nCould you please explain more about the 1st option?

", "votes": null, "replies": [ { "id": 1786087, "author_name": "ruchi798", "author_url": "", "post_date": "05/12/2022 15:42:26", "content": "

Hellooo!

\n

Docly automatically generates comments for the code blocks that we write.

\n

It can be installed using:
\npip install docly

", "votes": null, "replies": [] } ] }, { "id": 1786539, "author_name": "illgamhoduck", "author_url": "", "post_date": "05/13/2022 02:08:05", "content": "

Just for curiosity, is data generated from copilot safe from license problems?

", "votes": null, "replies": [] }, { "id": 1786616, "author_name": "muhammedtausif", "author_url": "", "post_date": "05/13/2022 04:15:28", "content": "

Docly is awesome. Thanks for sharing @ruchi798

", "votes": null, "replies": [] }, { "id": 1786998, "author_name": "atcelen", "author_url": "", "post_date": "05/13/2022 13:58:36", "content": "

Thanks for sharing!! Just wondering, is it inside the competition rules to use external data?

", "votes": null, "replies": [] }, { "id": 1789580, "author_name": "cv13j0", "author_url": "", "post_date": "05/14/2022 03:06:37", "content": "

Hello, @ruchi798 thanks for sharing; this is super useful, even outside of this competition.

", "votes": null, "replies": [] }, { "id": 1790361, "author_name": "dsxavier", "author_url": "", "post_date": "05/14/2022 20:17:56", "content": "

Is Dockly generating DocString you mean?

", "votes": null, "replies": [] }, { "id": 1791543, "author_name": "vidiqlogy", "author_url": "", "post_date": "05/16/2022 05:52:02", "content": "

I was wondering how we can use Docly in generating data ….

", "votes": null, "replies": [] }, { "id": 1792010, "author_name": "fajarwibowo", "author_url": "", "post_date": "05/16/2022 14:55:05", "content": "

good work, thanks for sharing

", "votes": null, "replies": [] }, { "id": 1792180, "author_name": "rcshubhadeep", "author_url": "", "post_date": "05/16/2022 18:30:40", "content": "

Hi @ruchi798 Thanks for the mentions. I created Docly three years ago. It was a very early stage product I am happy to see people still remembers it :) I am also thinking to take part in this competition.

", "votes": null, "replies": [ { "id": 1792207, "author_name": "ruchi798", "author_url": "", "post_date": "05/16/2022 18:46:57", "content": "

So nice to hear from you @rcshubhadeep 🎉
\nAll the best for the competition!

", "votes": null, "replies": [] } ] }, { "id": 1793476, "author_name": "arti1117", "author_url": "", "post_date": "05/18/2022 00:07:33", "content": "

Thank you for sharing

", "votes": null, "replies": [] }, { "id": 1794138, "author_name": "jaredfeng", "author_url": "", "post_date": "05/18/2022 14:24:48", "content": "

Another awesome idea:

\n

https://thisaidoesnotexist.com/

", "votes": null, "replies": [] }, { "id": 1794307, "author_name": "priyark", "author_url": "", "post_date": "05/18/2022 16:54:26", "content": "

Useful tools, I was knowing none of them.
\nThanks for sharing

", "votes": null, "replies": [] }, { "id": 1794375, "author_name": "tareknahool", "author_url": "", "post_date": "05/18/2022 18:25:59", "content": "

Good work for that job

", "votes": null, "replies": [] }, { "id": 1794749, "author_name": "vatsalsavaliya", "author_url": "", "post_date": "05/19/2022 06:07:05", "content": "

Thanks for sharing !

", "votes": null, "replies": [] }, { "id": 1800229, "author_name": "gazu468", "author_url": "", "post_date": "05/24/2022 19:19:27", "content": "

good work, thanks for sharing

", "votes": null, "replies": [] }, { "id": 1804639, "author_name": "paritoshmahto", "author_url": "", "post_date": "05/29/2022 09:52:16", "content": "

Thank you for sharing, Is GPT models work better in these type use cases?

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1785608": "New data can be generated for training using the following approaches:\n## 1) Code -> Comments 📚\n![](https://i.imgur.com/TgyH4vb.png)\nWe can use [Docly](http://thedocly.io/) to automatically generate the comments for our Python code.\n\n## 2) Comments -> Code 🚀\n![](https://i.imgur.com/ivYQLyZ.jpg)\nGithub Copilot can aid us in generating code for the code description we provide.\n\n## 🌟 Bonus: External data (?)\nAdditionally, there was a challenge hosted on Github a few years ago called the [CodeSearchNet challenge](https://github.blog/2019-09-26-introducing-the-codesearchnet-challenge/) which was based on Semantic Code Search.\n\n![](https://i.imgur.com/mor3WpH.png)\n\nThe [fully preprocessed Corpus for Python code](https://s3.amazonaws.com/code-search-net/CodeSearchNet/v2/python.zip) is available for download on Amazon S3. It contains about **6 million functions**. \n\nFeel free to share any ideas that you might have 🔮", "1786035": "ruchi798 Thank you.\nCould you please explain more about the 1st option?", "1786087": "Hellooo!\n\nDocly automatically generates comments for the code blocks that we write.\n\nIt can be installed using: \n`pip install docly`", "1786539": "Just for curiosity, is data generated from copilot safe from license problems?", "1786616": "**Docly** is awesome. Thanks for sharing @ruchi798", "1786998": "Thanks for sharing!! Just wondering, is it inside the competition rules to use external data?", "1789580": "Hello, @ruchi798 thanks for sharing; this is super useful, even outside of this competition.", "1790361": "Is **Dockly** generating DocString you mean?", "1791543": "I was wondering how we can use Docly in generating data ....", "1792010": "good work, thanks for sharing", "1792180": "Hi @ruchi798 Thanks for the mentions. I created Docly three years ago. It was a very early stage product I am happy to see people still remembers it :) I am also thinking to take part in this competition.", "1792207": "So nice to hear from you @rcshubhadeep 🎉\nAll the best for the competition!", "1793476": "Thank you for sharing", "1794138": "Another awesome idea:\n\nhttps://thisaidoesnotexist.com/", "1794307": "Useful tools, I was knowing none of them.\nThanks for sharing", "1794375": "Good work for that job", "1794749": "Thanks for sharing !", "1800229": "good work, thanks for sharing", "1804639": "Thank you for sharing, Is GPT models work better in these type use cases?" }, "source": "meta" }