I have found several datasets that seem to be useful for this competition:
\nHowever, I am not sure if it is permissible to use these datasets for training in accordance with the rules of the competition. Could you please clarify if training with these datasets is allowed by the rules?
\nThank you.
I have found several datasets that seem to be useful for this competition:
\nHowever, I am not sure if it is permissible to use these datasets for training in accordance with the rules of the competition. Could you please clarify if training with these datasets is allowed by the rules?
\nThank you.
Hey @kashiwaba ,
\nin Overview -> Code Requirements:
\n\n\nFreely & publicly available external data is allowed, including pre-trained models
\n
as far as I can tell, the first one has no license specified (should you contact the owner?) and the second one is licensed under the GNU General Public License v3.0
The rules allow external data in general.
\n\n\nC. External Data. You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).
\n
The licenses might be tricky though.
", "rawMarkdown": "The rules allow external data in general.\n\n> C. External Data. You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).\n\nThe licenses might be tricky though.", "votes": null }, { "id": "2230339", "postDate": "04/22/2023 09:12:37", "content": "Thank you all for your responses.
\nI apologize that my question was a bit unclear.
As you all have mentioned, I am wondering if it is always ok to use external datasets independent of the license.
\nFor example, it seems that at least for the source code of the solution, one cannot use GPL-licensed public code such as YOLOv5(Reference link), but would the same treatment apply to the dataset?
\nAlso, if the license is not clearly stated, is the use permitted? If so, is it necessary to ask the publisher for permission to use it?
\nI think this clarification of rules is very important to prevent participants from wasting time on unnecessary tasks.
\nIf anyone knows of any clear rules regarding the licensing of such usable external datasets, we would appreciate it if you would let us know.
@kashiwaba interesting links. Thank you.
", "rawMarkdown": "kashiwaba interesting links. Thank you.", "votes": null } ], "comments": [ { "id": 2230140, "author_name": "georgiyfedorov", "author_url": "", "post_date": "04/22/2023 04:00:44", "content": "Hey @kashiwaba ,
\nin Overview -> Code Requirements:
\n\n\nFreely & publicly available external data is allowed, including pre-trained models
\n
as far as I can tell, the first one has no license specified (should you contact the owner?) and the second one is licensed under the GNU General Public License v3.0
The rules allow external data in general.
\n\n\nC. External Data. You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).
\n
The licenses might be tricky though.
", "votes": null, "replies": [] }, { "id": 2230339, "author_name": "kashiwaba", "author_url": "", "post_date": "04/22/2023 09:12:37", "content": "Thank you all for your responses.
\nI apologize that my question was a bit unclear.
As you all have mentioned, I am wondering if it is always ok to use external datasets independent of the license.
\nFor example, it seems that at least for the source code of the solution, one cannot use GPL-licensed public code such as YOLOv5(Reference link), but would the same treatment apply to the dataset?
\nAlso, if the license is not clearly stated, is the use permitted? If so, is it necessary to ask the publisher for permission to use it?
\nI think this clarification of rules is very important to prevent participants from wasting time on unnecessary tasks.
\nIf anyone knows of any clear rules regarding the licensing of such usable external datasets, we would appreciate it if you would let us know.
@kashiwaba interesting links. Thank you.
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "2230078": "I have found several datasets that seem to be useful for this competition:\n- Chart-to-Text ([https://github.com/JasonObeid/Chart2Text](https://github.com/JasonObeid/Chart2Text))\n- ChartQA ([https://github.com/vis-nlp/ChartQA](https://github.com/vis-nlp/ChartQA))\n\nHowever, I am not sure if it is permissible to use these datasets for training in accordance with the rules of the competition. Could you please clarify if training with these datasets is allowed by the rules?\nThank you.", "2230140": "Hey @kashiwaba ,\n\nin Overview -> Code Requirements: \n>Freely & publicly available external data is allowed, including pre-trained models\n\nas far as I can tell, the first one has no license specified (should you contact the owner?) and the second one is licensed under the `GNU General Public License v3.0`", "2230313": "The rules allow external data in general.\n\n> C. External Data. You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).\n\nThe licenses might be tricky though.", "2230339": "Thank you all for your responses.\nI apologize that my question was a bit unclear.\n\nAs you all have mentioned, I am wondering if it is always ok to use external datasets independent of the license.\n\nFor example, it seems that at least for the source code of the solution, one cannot use GPL-licensed public code such as YOLOv5([Reference link](https://www.kaggle.com/competitions/benetech-making-graphs-accessible/discussion/396773#2208146)), but would the same treatment apply to the dataset? \n\nAlso, if the license is not clearly stated, is the use permitted? If so, is it necessary to ask the publisher for permission to use it?\n\nI think this clarification of rules is very important to prevent participants from wasting time on unnecessary tasks.\nIf anyone knows of any clear rules regarding the licensing of such usable external datasets, we would appreciate it if you would let us know.", "2230440": "kashiwaba interesting links. Thank you." }, "source": "meta" }