Please post requests for use of external data to this thread.
\n\nAs discussed in the pre-trained nets thread, this competition will allow limited use of external data, such as pre-trained nets. We will follow up with more specific rules and process on Monday, but for now our main considerations are 1) licensing (contestants and Yelp are freely able to use the data), 2) fairness (everyone has equal access to the data), 3) esstentialness (we want to see contestants' ingenuity, not scripting around an entire pre-trained model). Please expect the approval process to take a couple of business days while we learn about and discuss requests.
", "messages": [ { "id": "103397", "postDate": "01/02/2016 06:09:52", "content": "Please post requests for use of external data to this thread.
\n\nAs discussed in the pre-trained nets thread, this competition will allow limited use of external data, such as pre-trained nets. We will follow up with more specific rules and process on Monday, but for now our main considerations are 1) licensing (contestants and Yelp are freely able to use the data), 2) fairness (everyone has equal access to the data), 3) esstentialness (we want to see contestants' ingenuity, not scripting around an entire pre-trained model). Please expect the approval process to take a couple of business days while we learn about and discuss requests.
", "rawMarkdown": "Please post requests for use of external data to this thread.\r\n\r\nAs discussed in the pre-trained nets thread, this competition will allow limited use of external data, such as pre-trained nets. We will follow up with more specific rules and process on Monday, but for now our main considerations are 1) licensing (contestants and Yelp are freely able to use the data), 2) fairness (everyone has equal access to the data), 3) esstentialness (we want to see contestants' ingenuity, not scripting around an entire pre-trained model). Please expect the approval process to take a couple of business days while we learn about and discuss requests.", "votes": null }, { "id": "103744", "postDate": "01/06/2016 03:13:41", "content": "Are 'caffe' and the 'pretrained BVLC reference models under the BSD 2-Clause license' fair game?\nhttps://github.com/BVLC/caffe/
", "rawMarkdown": "Are 'caffe' and the 'pretrained BVLC reference models under the BSD 2-Clause license' fair game?\r\nhttps://github.com/BVLC/caffe/", "votes": null }, { "id": "103915", "postDate": "01/07/2016 22:43:12", "content": "@rbroberg You are free to use the Caffe models. Thanks for posting the request.
", "rawMarkdown": "rbroberg You are free to use the Caffe models. Thanks for posting the request.", "votes": null }, { "id": "107757", "postDate": "02/12/2016 10:02:00", "content": "Hi!
\n\nCan I use this VGG16 pretrained model for Keras?
\n\nhttps://gist.github.com/baraldilorenzo/07d7802847aaad0a35d3
\n\nKindly regards, Dmitry
", "rawMarkdown": "Hi!\r\n\r\nCan I use this VGG16 pretrained model for Keras?\r\n\r\nhttps://gist.github.com/baraldilorenzo/07d7802847aaad0a35d3\r\n\r\nKindly regards, Dmitry", "votes": null }, { "id": "107786", "postDate": "02/12/2016 16:37:24", "content": "Sure. Thanks for posting the request.
", "rawMarkdown": "Sure. Thanks for posting the request.", "votes": null }, { "id": "107854", "postDate": "02/13/2016 04:30:36", "content": "can we use noLearn ?
", "rawMarkdown": "can we use noLearn ?", "votes": null }, { "id": "107856", "postDate": "02/13/2016 04:47:56", "content": "https://github.com/dnouri/nolearn
", "rawMarkdown": "https://github.com/dnouri/nolearn", "votes": null }, { "id": "107861", "postDate": "02/13/2016 06:52:50", "content": "1: Can we use "ilsvrc12 image mean file" for caffe models? (we can download with https://github.com/BVLC/caffe/blob/master/data/ilsvrc12/get_ilsvrc_aux.sh)
\n\n2:Can we use "channel-wise mean value of ilsvrc12 dataset"? (Values of 18~20th-rows of https://github.com/BVLC/caffe/blob/master/models/bvlc_reference_caffenet/train_val.prototxt)
\n\n3:You said we can use VGG16 model but "https://gist.github.com/ksimonyan/211839e770f7b538e2d8#file-readme-md" is saying "non-commercial use only", is it OK to use VGG16 model in this competition?
", "rawMarkdown": "1: Can we use \"ilsvrc12 image mean file\" for caffe models? (we can download with https://github.com/BVLC/caffe/blob/master/data/ilsvrc12/get_ilsvrc_aux.sh)\r\n\r\n2:Can we use \"channel-wise mean value of ilsvrc12 dataset\"? (Values of 18~20th-rows of https://github.com/BVLC/caffe/blob/master/models/bvlc_reference_caffenet/train_val.prototxt)\r\n\r\n3:You said we can use VGG16 model but \"https://gist.github.com/ksimonyan/211839e770f7b538e2d8#file-readme-md\" is saying \"non-commercial use only\", is it OK to use VGG16 model in this competition?", "votes": null }, { "id": "107863", "postDate": "02/13/2016 07:11:28", "content": "@ki thanks for bringing this up. Seems like using VGG16 is against the rule since Yelp has no right to use it due to the non-commercial license.
\n\nnolearn and the caffe data is OK to use.
", "rawMarkdown": "ki thanks for bringing this up. Seems like using VGG16 is against the rule since Yelp has no right to use it due to the non-commercial license.\r\n\r\nnolearn and the caffe data is OK to use.", "votes": null }, { "id": "107950", "postDate": "02/14/2016 03:27:21", "content": "Can we use theano, torch, tensor?
", "rawMarkdown": "Can we use theano, torch, tensor?", "votes": null }, { "id": "107951", "postDate": "02/14/2016 03:27:36", "content": "", "rawMarkdown": "", "votes": null }, { "id": "108368", "postDate": "02/16/2016 21:59:30", "content": "You are free to use open-sourced packages that allow commercial usage. theano, torch, tensorflow all belongs to this category
", "rawMarkdown": "You are free to use open-sourced packages that allow commercial usage. theano, torch, tensorflow all belongs to this category", "votes": null }, { "id": "109071", "postDate": "02/23/2016 00:43:04", "content": "can we use: https://github.com/soumith/inception.torch
", "rawMarkdown": "can we use: https://github.com/soumith/inception.torch", "votes": null }, { "id": "109296", "postDate": "02/24/2016 21:14:20", "content": "Yes you can use the inception net
", "rawMarkdown": "Yes you can use the inception net", "votes": null }, { "id": "110204", "postDate": "03/03/2016 18:59:58", "content": "OK, so I'm a little confused. Here http://www.robots.ox.ac.uk/~vgg/research/very_deep/ there is information that VGG16 and VGG19 are under CC4 with commercial usage
", "rawMarkdown": "OK, so I'm a little confused. Here http://www.robots.ox.ac.uk/~vgg/research/very_deep/ there is information that VGG16 and VGG19 are under CC4 with commercial usage", "votes": null }, { "id": "110207", "postDate": "03/03/2016 19:25:14", "content": "But the pages\nhttps://gist.github.com/ksimonyan/3785162f95cd2d5fee77#file-readme-md\nhttps://gist.github.com/ksimonyan/211839e770f7b538e2d8#file-readme-md
\n\nSuggests that they are non-commercial. Seems like it is not very consistent here.
\n\nIf you are very interested in trying these VGG models please confirm with the original authors what the actual license they are using and have them update the pages. Otherwise please avoid using them for the current contest. Thanks!
", "rawMarkdown": "But the pages\r\nhttps://gist.github.com/ksimonyan/3785162f95cd2d5fee77#file-readme-md\r\nhttps://gist.github.com/ksimonyan/211839e770f7b538e2d8#file-readme-md\r\n\r\nSuggests that they are non-commercial. Seems like it is not very consistent here.\r\n\r\nIf you are very interested in trying these VGG models please confirm with the original authors what the actual license they are using and have them update the pages. Otherwise please avoid using them for the current contest. Thanks!", "votes": null }, { "id": "110859", "postDate": "03/08/2016 22:36:54", "content": "Are all Caffe models in the model zoo allowed? Are we allowed to use prebuilt or pre-trained torch models too?
", "rawMarkdown": "Are all Caffe models in the model zoo allowed? Are we allowed to use prebuilt or pre-trained torch models too?", "votes": null }, { "id": "110863", "postDate": "03/08/2016 23:53:49", "content": "Not necessarily. The model needs to be free for commercial usage, which does not apply to certain licenses (e.g. GPL). While most models should be OK, if you are interested a certain model, please post it here and we will check the license.
", "rawMarkdown": "Not necessarily. The model needs to be free for commercial usage, which does not apply to certain licenses (e.g. GPL). While most models should be OK, if you are interested a certain model, please post it here and we will check the license.", "votes": null }, { "id": "112546", "postDate": "03/21/2016 21:18:12", "content": "[quote=Fang-Chieh Chou;110863]
\n\nNot necessarily. The model needs to be free for commercial usage, which does not apply to certain licenses (e.g. GPL). While most models should be OK, if you are interested a certain model, please post it here and we will check the license.
\n\n[/quote]
\n\nI am sorry, what commercial usage are you talking about? This is a Recruiting Competition, and two other admins confirmed this is about hiring and job skills only!
\n\n[quote]RECRUITING COMPETITION SUBMISSION LICENSE GRANT
\n\nIf the Competition is classed above as a Recruiting Competition, Participant agrees to grant and hereby grants to Competition Sponsor a perpetual, irrevocable, worldwide, royalty-free, transferable, and sublicenseable right to use, review, reproduce, and internally distribute the Submission (including any and all submitted source code) in connection with evaluating Participant's suitability for employment, provided that the Submission licensed as set forth above will not be used for Competition Host’s commercial purposes except as may otherwise be set forth in this Agreement.[/quote]
\n\n[quote=ody;110344]\nJohn, we are not planning to use the winning solution commercially; this is mostly a hiring exercise, a chance to share an interesting/less-common type of machine learning problem (and dataset) with the Kaggle community, and for academic interest. [/quote]
\n\n[quote=Jim B;108328]\n This contest is a chance to let the results speak for themselves -- if you can create great algorithms, then it doesn't matter what you have on your resume, we want to talk to you.
\n\nReleasing features at Yelp is about more than writing an algorithm alone with supplied data and one metric, so we need to test for the other abilities as well. That means we can't guarantee a job offer.\n[/quote]
", "rawMarkdown": "[quote=Fang-Chieh Chou;110863]\r\n\r\nNot necessarily. The model needs to be free for commercial usage, which does not apply to certain licenses (e.g. GPL). While most models should be OK, if you are interested a certain model, please post it here and we will check the license.\r\n\r\n[/quote]\r\n\r\n\r\nI am sorry, what commercial usage are you talking about? This is a Recruiting Competition, and two other admins confirmed this is about hiring and job skills only!\r\n\r\n\r\n[quote]RECRUITING COMPETITION SUBMISSION LICENSE GRANT\r\n\r\nIf the Competition is classed above as a Recruiting Competition, Participant agrees to grant and hereby grants to Competition Sponsor a perpetual, irrevocable, worldwide, royalty-free, transferable, and sublicenseable right to use, review, reproduce, and internally distribute the Submission (including any and all submitted source code) in connection with evaluating Participant's suitability for employment, provided that the Submission licensed as set forth above will **not be used for Competition Host’s commercial purposes** except as may otherwise be set forth in this Agreement.[/quote]\r\n\r\n[quote=ody;110344]\r\nJohn, we are not planning to use the winning solution commercially; this is mostly a hiring exercise, a chance to share an interesting/less-common type of machine learning problem (and dataset) with the Kaggle community, and for academic interest. [/quote]\r\n\r\n[quote=Jim B;108328]\r\n This contest is a chance to let the results speak for themselves -- if you can create great algorithms, then it doesn't matter what you have on your resume, we want to talk to you.\r\n\r\nReleasing features at Yelp is about more than writing an algorithm alone with supplied data and one metric, so we need to test for the other abilities as well. That means we can't guarantee a job offer.\r\n[/quote]", "votes": null }, { "id": "112553", "postDate": "03/21/2016 22:50:40", "content": "@rakhlin Thanks for asking, but the current rule also states that
\n\nOPEN-SOURCE CODE\n\nA Submission will be ineligible to win a prize if it was developed using code containing or depending on software licensed under an open source license:\n\n* other than an Open Source Initiative-approved license (see <http://opensource.org/>); or\n* an open source license that prohibits commercial use.\nIndeed this competition is about skill and hiring only and we don't have a plan on using them for commercial purposes, but we have to respect the rule of Kaggle as well. Therefore please refrain from using noncommercial-only external data. The rule you cites is about your agreement to the IP of your code submitted to Kaggle, and the rule for the winning submission is stricter.
", "rawMarkdown": "rakhlin Thanks for asking, but the current rule also states that\r\n\r\n OPEN-SOURCE CODE\r\n \r\n A Submission will be ineligible to win a prize if it was developed using code containing or depending on software licensed under an open source license:\r\n \r\n * other than an Open Source Initiative-approved license (see@Fang-Chieh Chou could you please clarify what prizes you are referring to speaking of this competition? It seems, the rule you citing specifically applies to competitions giving out prizes, and this one is different.
", "rawMarkdown": "Fang-Chieh Chou could you please clarify what prizes you are referring to speaking of this competition? It seems, the rule you citing specifically applies to competitions giving out prizes, and this one is different.", "votes": null }, { "id": "112573", "postDate": "03/22/2016 01:50:49", "content": "@rakhlin In this case the prize will be the chance to get fast-track interview with Yelp, and possibly a post on Yelp's blog to highlight the winner's story and profile (we are still discussing the details for the blog post).
\n\nThe first post in this thread also states that the license of external data must allow free usage by contestant and Yelp. While we don't plan using your winning code for commercial, we might want to test/check your code, and the noncommercial license will make it legally complicated.
\n\nI think this is not a significant restraint since most of the tools/data out there use valid licenses.
", "rawMarkdown": "rakhlin In this case the prize will be the chance to get fast-track interview with Yelp, and possibly a post on Yelp's blog to highlight the winner's story and profile (we are still discussing the details for the blog post).\r\n\r\nThe first post in this thread also states that the license of external data must allow free usage by contestant and Yelp. While we don't plan using your winning code for commercial, we might want to test/check your code, and the noncommercial license will make it legally complicated.\r\n\r\nI think this is not a significant restraint since most of the tools/data out there use valid licenses.", "votes": null }, { "id": "112595", "postDate": "03/22/2016 08:00:59", "content": "@Fang-Chieh Chou, I am pretty sure that the chance of getting fast-track interview, just like any chance overall, is not a 'prize' from a legal point of view. Prize is something unconditional (and more or less material). In my opinion, if Yelp is serious about not using this work for business purposes, licenses shouldn't matter.
", "rawMarkdown": "Fang-Chieh Chou, I am pretty sure that the chance of getting fast-track interview, just like any chance overall, is not a 'prize' from a legal point of view. Prize is something unconditional (and more or less material). In my opinion, if Yelp is serious about not using this work for business purposes, licenses shouldn't matter.", "votes": null }, { "id": "112657", "postDate": "03/22/2016 17:02:48", "content": "@rakhlin, while your argument makes sense, I am no law expert so I cannot be sure whether non-commercial license is allowed given the current rule. Since we have been following this rule interpretation for the majority of the contest I think it is better not to change it at the near-final stage.
\n\nAs I stated above I think the restraint here is fairly liberal and you should be able to use most of the existing tools out there. Unless you still have a strong opinion on this, please refrain from using non-commercially licensed data. Thanks!
", "rawMarkdown": "rakhlin, while your argument makes sense, I am no law expert so I cannot be sure whether non-commercial license is allowed given the current rule. Since we have been following this rule interpretation for the majority of the contest I think it is better not to change it at the near-final stage.\r\n\r\nAs I stated above I think the restraint here is fairly liberal and you should be able to use most of the existing tools out there. Unless you still have a strong opinion on this, please refrain from using non-commercially licensed data. Thanks!", "votes": null }, { "id": "112996", "postDate": "03/25/2016 23:34:21", "content": "I am not going to participate, but think that this models could help :)\nhttps://github.com/gcr/torch-residual-networks
\n\nhttps://github.com/gcr/torch-residual-networks/blob/master/LICENSE.md\nLicense:
\n\n\n\n\nCopyright (c) 2016 Michael Wilber
\n \nThis software is provided 'as-is', without any express or implied\n warranty. In no event will the authors be held liable for any damages\n arising from the use of this software.
\n \nPermission is granted to anyone to use this software for any purpose,\n including commercial applications, and to alter it and redistribute it\n freely, subject to the following restrictions:
\n \n\n
\n- The origin of this software must not be misrepresented; you must not claim that you wrote the original software. If you use this\n software in a product, an acknowledgement in the product\n documentation would be appreciated but is not required.
\n- Altered source versions must be plainly marked as such, and must not be misrepresented as being the original software.
\n- This notice may not be removed or altered from any source distribution.
\n
And caffe one:
\n\nhttps://github.com/KaimingHe/deep-residual-networks
\n\n\n", "rawMarkdown": "I am not going to participate, but think that this models could help :)\r\nhttps://github.com/gcr/torch-residual-networks\r\n \r\nhttps://github.com/gcr/torch-residual-networks/blob/master/LICENSE.md\r\nLicense:\r\n\r\n> Copyright (c) 2016 Michael Wilber\r\n> \r\n> This software is provided 'as-is', without any express or implied\r\n> warranty. In no event will the authors be held liable for any damages\r\n> arising from the use of this software.\r\n> \r\n> Permission is granted to anyone to use this software for any purpose,\r\n> including commercial applications, and to alter it and redistribute it\r\n> freely, subject to the following restrictions:\r\n> \r\n> 1. The origin of this software must not be misrepresented; you must not claim that you wrote the original software. If you use this\r\n> software in a product, an acknowledgement in the product\r\n> documentation would be appreciated but is not required.\r\n> 2. Altered source versions must be plainly marked as such, and must not be misrepresented as being the original software.\r\n> 3. This notice may not be removed or altered from any source distribution.\r\n\r\nAnd caffe one:\r\n\r\nhttps://github.com/KaimingHe/deep-residual-networks\r\n\r\n> The MIT License (MIT)\r\n> \r\n> Copyright (c) 2016 Shaoqing Ren\r\n> \r\n> Permission is hereby granted, free of charge, to any person obtaining\r\n> a copy of this software and associated documentation files (the\r\n> \"Software\"), to deal in the Software without restriction, including\r\n> without limitation the rights to use, copy, modify, merge, publish,\r\n> distribute, sublicense, and/or sell copies of the Software, and to\r\n> permit persons to whom the Software is furnished to do so, subject to\r\n> the following conditions:\r\n> \r\n> The above copyright notice and this permission notice shall be\r\n> included in all copies or substantial portions of the Software.\r\n> \r\n> THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND,\r\n> EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF\r\n> MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.\r\n> IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY\r\n> CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,\r\n> TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE\r\n> SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.", "votes": null }, { "id": "113423", "postDate": "03/31/2016 16:28:05", "content": "The MIT License (MIT)
\n \nCopyright (c) 2016 Shaoqing Ren
\n \nPermission is hereby granted, free of charge, to any person obtaining\n a copy of this software and associated documentation files (the\n "Software"), to deal in the Software without restriction, including\n without limitation the rights to use, copy, modify, merge, publish,\n distribute, sublicense, and/or sell copies of the Software, and to\n permit persons to whom the Software is furnished to do so, subject to\n the following conditions:
\n \nThe above copyright notice and this permission notice shall be\n included in all copies or substantial portions of the Software.
\n \nTHE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,\n EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF\n MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.\n IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY\n CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,\n TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE\n SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
\n
To competition admin:\nI am using this residual networks. Could you please confirm it is allowed or not?
\n\n[quote=old-ufo;112996]
\n\nI am not going to participate, but think that this models could help :)\nhttps://github.com/gcr/torch-residual-networks
\n\nhttps://github.com/gcr/torch-residual-networks/blob/master/LICENSE.md\nLicense:
\n\n\n\n\nCopyright (c) 2016 Michael Wilber
\n \nThis software is provided 'as-is', without any express or implied\n warranty. In no event will the authors be held liable for any damages\n arising from the use of this software.
\n \nPermission is granted to anyone to use this software for any purpose,\n including commercial applications, and to alter it and redistribute it\n freely, subject to the following restrictions:
\n \n\n
\n- The origin of this software must not be misrepresented; you must not claim that you wrote the original software. If you use this\n software in a product, an acknowledgement in the product\n documentation would be appreciated but is not required.
\n- Altered source versions must be plainly marked as such, and must not be misrepresented as being the original software.
\n- This notice may not be removed or altered from any source distribution.
\n
And caffe one:
\n\nhttps://github.com/KaimingHe/deep-residual-networks
\n\n\n\n\nThe MIT License (MIT)
\n \nCopyright (c) 2016 Shaoqing Ren
\n \nPermission is hereby granted, free of charge, to any person obtaining\n a copy of this software and associated documentation files (the\n "Software"), to deal in the Software without restriction, including\n without limitation the rights to use, copy, modify, merge, publish,\n distribute, sublicense, and/or sell copies of the Software, and to\n permit persons to whom the Software is furnished to do so, subject to\n the following conditions:
\n \nThe above copyright notice and this permission notice shall be\n included in all copies or substantial portions of the Software.
\n \nTHE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,\n EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF\n MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.\n IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY\n CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,\n TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE\n SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
\n
[/quote]
", "rawMarkdown": "To competition admin:\r\nI am using this residual networks. Could you please confirm it is allowed or not?\r\n\r\n[quote=old-ufo;112996]\r\n\r\nI am not going to participate, but think that this models could help :)\r\nhttps://github.com/gcr/torch-residual-networks\r\n \r\nhttps://github.com/gcr/torch-residual-networks/blob/master/LICENSE.md\r\nLicense:\r\n\r\n> Copyright (c) 2016 Michael Wilber\r\n> \r\n> This software is provided 'as-is', without any express or implied\r\n> warranty. In no event will the authors be held liable for any damages\r\n> arising from the use of this software.\r\n> \r\n> Permission is granted to anyone to use this software for any purpose,\r\n> including commercial applications, and to alter it and redistribute it\r\n> freely, subject to the following restrictions:\r\n> \r\n> 1. The origin of this software must not be misrepresented; you must not claim that you wrote the original software. If you use this\r\n> software in a product, an acknowledgement in the product\r\n> documentation would be appreciated but is not required.\r\n> 2. Altered source versions must be plainly marked as such, and must not be misrepresented as being the original software.\r\n> 3. This notice may not be removed or altered from any source distribution.\r\n\r\nAnd caffe one:\r\n\r\nhttps://github.com/KaimingHe/deep-residual-networks\r\n\r\n> The MIT License (MIT)\r\n> \r\n> Copyright (c) 2016 Shaoqing Ren\r\n> \r\n> Permission is hereby granted, free of charge, to any person obtaining\r\n> a copy of this software and associated documentation files (the\r\n> \"Software\"), to deal in the Software without restriction, including\r\n> without limitation the rights to use, copy, modify, merge, publish,\r\n> distribute, sublicense, and/or sell copies of the Software, and to\r\n> permit persons to whom the Software is furnished to do so, subject to\r\n> the following conditions:\r\n> \r\n> The above copyright notice and this permission notice shall be\r\n> included in all copies or substantial portions of the Software.\r\n> \r\n> THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND,\r\n> EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF\r\n> MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.\r\n> IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY\r\n> CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,\r\n> TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE\r\n> SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.\r\n\r\n[/quote]", "votes": null }, { "id": "113456", "postDate": "04/01/2016 04:38:23", "content": "Please confirm if the mxnet pretrained nets mentioned here are allowed. The licensing information is provided here
\n\nThanks in advance.
", "rawMarkdown": "Please confirm if the mxnet pretrained nets mentioned [here][1] are allowed. The licensing information is provided [here][2]\r\n\r\nThanks in advance.\r\n\r\n [1]: https://www.kaggle.com/c/yelp-restaurant-photo-classification/forums/t/18199/external-data-mxnet-pretrained-model/103326#post103326\r\n [2]: https://www.kaggle.com/c/yelp-restaurant-photo-classification/forums/t/18199/external-data-mxnet-pretrained-model/103396#post103396", "votes": null }, { "id": "113474", "postDate": "04/01/2016 08:29:27", "content": "Hi, please confirm that we can use the MIT models listed here: http://places.csail.mit.edu/downloadCNN.html
\n\nAlternatively, 3 of the 4 can be found listed here too: https://github.com/BVLC/caffe/wiki/Model-Zoo
\n\nNote that I'd also checked with the main contact listed on the website, and he got back that they can be used without restrictions, but would be great to give credit (eg link to project page - http://places.csail.mit.edu).
\n\nThanks
", "rawMarkdown": "Hi, please confirm that we can use the MIT models listed here: http://places.csail.mit.edu/downloadCNN.html\r\n\r\nAlternatively, 3 of the 4 can be found listed here too: https://github.com/BVLC/caffe/wiki/Model-Zoo\r\n\r\nNote that I'd also checked with the main contact listed on the website, and he got back that they can be used without restrictions, but would be great to give credit (eg link to project page - http://places.csail.mit.edu).\r\n\r\nThanks", "votes": null }, { "id": "113515", "postDate": "04/01/2016 15:32:31", "content": "@mango0211, @Prithvi, @Blue Light:\nYeah I took a look and the models look OK. Go ahead and use them
", "rawMarkdown": "mango0211, @Prithvi, @Blue Light:\r\nYeah I took a look and the models look OK. Go ahead and use them", "votes": null } ], "comments": [ { "id": 103744, "author_name": "rbroberg", "author_url": "", "post_date": "01/06/2016 03:13:41", "content": "Are 'caffe' and the 'pretrained BVLC reference models under the BSD 2-Clause license' fair game?\nhttps://github.com/BVLC/caffe/
", "votes": null, "replies": [] }, { "id": 103915, "author_name": "fcchou", "author_url": "", "post_date": "01/07/2016 22:43:12", "content": "@rbroberg You are free to use the Caffe models. Thanks for posting the request.
", "votes": null, "replies": [] }, { "id": 107757, "author_name": "deaddy", "author_url": "", "post_date": "02/12/2016 10:02:00", "content": "Hi!
\n\nCan I use this VGG16 pretrained model for Keras?
\n\nhttps://gist.github.com/baraldilorenzo/07d7802847aaad0a35d3
\n\nKindly regards, Dmitry
", "votes": null, "replies": [] }, { "id": 107786, "author_name": "fcchou", "author_url": "", "post_date": "02/12/2016 16:37:24", "content": "Sure. Thanks for posting the request.
", "votes": null, "replies": [] }, { "id": 107854, "author_name": "akalimil", "author_url": "", "post_date": "02/13/2016 04:30:36", "content": "can we use noLearn ?
", "votes": null, "replies": [] }, { "id": 107856, "author_name": "akalimil", "author_url": "", "post_date": "02/13/2016 04:47:56", "content": "https://github.com/dnouri/nolearn
", "votes": null, "replies": [] }, { "id": 107861, "author_name": "hogefuga", "author_url": "", "post_date": "02/13/2016 06:52:50", "content": "1: Can we use "ilsvrc12 image mean file" for caffe models? (we can download with https://github.com/BVLC/caffe/blob/master/data/ilsvrc12/get_ilsvrc_aux.sh)
\n\n2:Can we use "channel-wise mean value of ilsvrc12 dataset"? (Values of 18~20th-rows of https://github.com/BVLC/caffe/blob/master/models/bvlc_reference_caffenet/train_val.prototxt)
\n\n3:You said we can use VGG16 model but "https://gist.github.com/ksimonyan/211839e770f7b538e2d8#file-readme-md" is saying "non-commercial use only", is it OK to use VGG16 model in this competition?
", "votes": null, "replies": [] }, { "id": 107863, "author_name": "fcchou", "author_url": "", "post_date": "02/13/2016 07:11:28", "content": "@ki thanks for bringing this up. Seems like using VGG16 is against the rule since Yelp has no right to use it due to the non-commercial license.
\n\nnolearn and the caffe data is OK to use.
", "votes": null, "replies": [] }, { "id": 107950, "author_name": "akalimil", "author_url": "", "post_date": "02/14/2016 03:27:21", "content": "Can we use theano, torch, tensor?
", "votes": null, "replies": [] }, { "id": 107951, "author_name": "akalimil", "author_url": "", "post_date": "02/14/2016 03:27:36", "content": "", "votes": null, "replies": [] }, { "id": 108368, "author_name": "fcchou", "author_url": "", "post_date": "02/16/2016 21:59:30", "content": "You are free to use open-sourced packages that allow commercial usage. theano, torch, tensorflow all belongs to this category
", "votes": null, "replies": [] }, { "id": 109071, "author_name": "faceplant", "author_url": "", "post_date": "02/23/2016 00:43:04", "content": "can we use: https://github.com/soumith/inception.torch
", "votes": null, "replies": [] }, { "id": 109296, "author_name": "fcchou", "author_url": "", "post_date": "02/24/2016 21:14:20", "content": "Yes you can use the inception net
", "votes": null, "replies": [] }, { "id": 110204, "author_name": "adamszalucha", "author_url": "", "post_date": "03/03/2016 18:59:58", "content": "OK, so I'm a little confused. Here http://www.robots.ox.ac.uk/~vgg/research/very_deep/ there is information that VGG16 and VGG19 are under CC4 with commercial usage
", "votes": null, "replies": [] }, { "id": 110207, "author_name": "fcchou", "author_url": "", "post_date": "03/03/2016 19:25:14", "content": "But the pages\nhttps://gist.github.com/ksimonyan/3785162f95cd2d5fee77#file-readme-md\nhttps://gist.github.com/ksimonyan/211839e770f7b538e2d8#file-readme-md
\n\nSuggests that they are non-commercial. Seems like it is not very consistent here.
\n\nIf you are very interested in trying these VGG models please confirm with the original authors what the actual license they are using and have them update the pages. Otherwise please avoid using them for the current contest. Thanks!
", "votes": null, "replies": [] }, { "id": 110859, "author_name": "adityagudimella", "author_url": "", "post_date": "03/08/2016 22:36:54", "content": "Are all Caffe models in the model zoo allowed? Are we allowed to use prebuilt or pre-trained torch models too?
", "votes": null, "replies": [] }, { "id": 110863, "author_name": "fcchou", "author_url": "", "post_date": "03/08/2016 23:53:49", "content": "Not necessarily. The model needs to be free for commercial usage, which does not apply to certain licenses (e.g. GPL). While most models should be OK, if you are interested a certain model, please post it here and we will check the license.
", "votes": null, "replies": [] }, { "id": 112546, "author_name": "rakhlin", "author_url": "", "post_date": "03/21/2016 21:18:12", "content": "[quote=Fang-Chieh Chou;110863]
\n\nNot necessarily. The model needs to be free for commercial usage, which does not apply to certain licenses (e.g. GPL). While most models should be OK, if you are interested a certain model, please post it here and we will check the license.
\n\n[/quote]
\n\nI am sorry, what commercial usage are you talking about? This is a Recruiting Competition, and two other admins confirmed this is about hiring and job skills only!
\n\n[quote]RECRUITING COMPETITION SUBMISSION LICENSE GRANT
\n\nIf the Competition is classed above as a Recruiting Competition, Participant agrees to grant and hereby grants to Competition Sponsor a perpetual, irrevocable, worldwide, royalty-free, transferable, and sublicenseable right to use, review, reproduce, and internally distribute the Submission (including any and all submitted source code) in connection with evaluating Participant's suitability for employment, provided that the Submission licensed as set forth above will not be used for Competition Host’s commercial purposes except as may otherwise be set forth in this Agreement.[/quote]
\n\n[quote=ody;110344]\nJohn, we are not planning to use the winning solution commercially; this is mostly a hiring exercise, a chance to share an interesting/less-common type of machine learning problem (and dataset) with the Kaggle community, and for academic interest. [/quote]
\n\n[quote=Jim B;108328]\n This contest is a chance to let the results speak for themselves -- if you can create great algorithms, then it doesn't matter what you have on your resume, we want to talk to you.
\n\nReleasing features at Yelp is about more than writing an algorithm alone with supplied data and one metric, so we need to test for the other abilities as well. That means we can't guarantee a job offer.\n[/quote]
", "votes": null, "replies": [] }, { "id": 112553, "author_name": "fcchou", "author_url": "", "post_date": "03/21/2016 22:50:40", "content": "@rakhlin Thanks for asking, but the current rule also states that
\n\nOPEN-SOURCE CODE\n\nA Submission will be ineligible to win a prize if it was developed using code containing or depending on software licensed under an open source license:\n\n* other than an Open Source Initiative-approved license (see <http://opensource.org/>); or\n* an open source license that prohibits commercial use.\nIndeed this competition is about skill and hiring only and we don't have a plan on using them for commercial purposes, but we have to respect the rule of Kaggle as well. Therefore please refrain from using noncommercial-only external data. The rule you cites is about your agreement to the IP of your code submitted to Kaggle, and the rule for the winning submission is stricter.
", "votes": null, "replies": [] }, { "id": 112572, "author_name": "rakhlin", "author_url": "", "post_date": "03/22/2016 01:40:09", "content": "@Fang-Chieh Chou could you please clarify what prizes you are referring to speaking of this competition? It seems, the rule you citing specifically applies to competitions giving out prizes, and this one is different.
", "votes": null, "replies": [] }, { "id": 112573, "author_name": "fcchou", "author_url": "", "post_date": "03/22/2016 01:50:49", "content": "@rakhlin In this case the prize will be the chance to get fast-track interview with Yelp, and possibly a post on Yelp's blog to highlight the winner's story and profile (we are still discussing the details for the blog post).
\n\nThe first post in this thread also states that the license of external data must allow free usage by contestant and Yelp. While we don't plan using your winning code for commercial, we might want to test/check your code, and the noncommercial license will make it legally complicated.
\n\nI think this is not a significant restraint since most of the tools/data out there use valid licenses.
", "votes": null, "replies": [] }, { "id": 112595, "author_name": "rakhlin", "author_url": "", "post_date": "03/22/2016 08:00:59", "content": "@Fang-Chieh Chou, I am pretty sure that the chance of getting fast-track interview, just like any chance overall, is not a 'prize' from a legal point of view. Prize is something unconditional (and more or less material). In my opinion, if Yelp is serious about not using this work for business purposes, licenses shouldn't matter.
", "votes": null, "replies": [] }, { "id": 112657, "author_name": "fcchou", "author_url": "", "post_date": "03/22/2016 17:02:48", "content": "@rakhlin, while your argument makes sense, I am no law expert so I cannot be sure whether non-commercial license is allowed given the current rule. Since we have been following this rule interpretation for the majority of the contest I think it is better not to change it at the near-final stage.
\n\nAs I stated above I think the restraint here is fairly liberal and you should be able to use most of the existing tools out there. Unless you still have a strong opinion on this, please refrain from using non-commercially licensed data. Thanks!
", "votes": null, "replies": [] }, { "id": 112996, "author_name": "oldufo", "author_url": "", "post_date": "03/25/2016 23:34:21", "content": "I am not going to participate, but think that this models could help :)\nhttps://github.com/gcr/torch-residual-networks
\n\nhttps://github.com/gcr/torch-residual-networks/blob/master/LICENSE.md\nLicense:
\n\n\n\n\nCopyright (c) 2016 Michael Wilber
\n \nThis software is provided 'as-is', without any express or implied\n warranty. In no event will the authors be held liable for any damages\n arising from the use of this software.
\n \nPermission is granted to anyone to use this software for any purpose,\n including commercial applications, and to alter it and redistribute it\n freely, subject to the following restrictions:
\n \n\n
\n- The origin of this software must not be misrepresented; you must not claim that you wrote the original software. If you use this\n software in a product, an acknowledgement in the product\n documentation would be appreciated but is not required.
\n- Altered source versions must be plainly marked as such, and must not be misrepresented as being the original software.
\n- This notice may not be removed or altered from any source distribution.
\n
And caffe one:
\n\nhttps://github.com/KaimingHe/deep-residual-networks
\n\n\n", "votes": null, "replies": [] }, { "id": 113423, "author_name": "mango0211", "author_url": "", "post_date": "03/31/2016 16:28:05", "content": "The MIT License (MIT)
\n \nCopyright (c) 2016 Shaoqing Ren
\n \nPermission is hereby granted, free of charge, to any person obtaining\n a copy of this software and associated documentation files (the\n "Software"), to deal in the Software without restriction, including\n without limitation the rights to use, copy, modify, merge, publish,\n distribute, sublicense, and/or sell copies of the Software, and to\n permit persons to whom the Software is furnished to do so, subject to\n the following conditions:
\n \nThe above copyright notice and this permission notice shall be\n included in all copies or substantial portions of the Software.
\n \nTHE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,\n EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF\n MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.\n IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY\n CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,\n TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE\n SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
\n
To competition admin:\nI am using this residual networks. Could you please confirm it is allowed or not?
\n\n[quote=old-ufo;112996]
\n\nI am not going to participate, but think that this models could help :)\nhttps://github.com/gcr/torch-residual-networks
\n\nhttps://github.com/gcr/torch-residual-networks/blob/master/LICENSE.md\nLicense:
\n\n\n\n\nCopyright (c) 2016 Michael Wilber
\n \nThis software is provided 'as-is', without any express or implied\n warranty. In no event will the authors be held liable for any damages\n arising from the use of this software.
\n \nPermission is granted to anyone to use this software for any purpose,\n including commercial applications, and to alter it and redistribute it\n freely, subject to the following restrictions:
\n \n\n
\n- The origin of this software must not be misrepresented; you must not claim that you wrote the original software. If you use this\n software in a product, an acknowledgement in the product\n documentation would be appreciated but is not required.
\n- Altered source versions must be plainly marked as such, and must not be misrepresented as being the original software.
\n- This notice may not be removed or altered from any source distribution.
\n
And caffe one:
\n\nhttps://github.com/KaimingHe/deep-residual-networks
\n\n\n\n\nThe MIT License (MIT)
\n \nCopyright (c) 2016 Shaoqing Ren
\n \nPermission is hereby granted, free of charge, to any person obtaining\n a copy of this software and associated documentation files (the\n "Software"), to deal in the Software without restriction, including\n without limitation the rights to use, copy, modify, merge, publish,\n distribute, sublicense, and/or sell copies of the Software, and to\n permit persons to whom the Software is furnished to do so, subject to\n the following conditions:
\n \nThe above copyright notice and this permission notice shall be\n included in all copies or substantial portions of the Software.
\n \nTHE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,\n EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF\n MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.\n IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY\n CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,\n TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE\n SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
\n
[/quote]
", "votes": null, "replies": [] }, { "id": 113456, "author_name": "prith189", "author_url": "", "post_date": "04/01/2016 04:38:23", "content": "Please confirm if the mxnet pretrained nets mentioned here are allowed. The licensing information is provided here
\n\nThanks in advance.
", "votes": null, "replies": [] }, { "id": 113474, "author_name": "bluelight773", "author_url": "", "post_date": "04/01/2016 08:29:27", "content": "Hi, please confirm that we can use the MIT models listed here: http://places.csail.mit.edu/downloadCNN.html
\n\nAlternatively, 3 of the 4 can be found listed here too: https://github.com/BVLC/caffe/wiki/Model-Zoo
\n\nNote that I'd also checked with the main contact listed on the website, and he got back that they can be used without restrictions, but would be great to give credit (eg link to project page - http://places.csail.mit.edu).
\n\nThanks
", "votes": null, "replies": [] }, { "id": 113515, "author_name": "fcchou", "author_url": "", "post_date": "04/01/2016 15:32:31", "content": "@mango0211, @Prithvi, @Blue Light:\nYeah I took a look and the models look OK. Go ahead and use them
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "103397": "Please post requests for use of external data to this thread.\r\n\r\nAs discussed in the pre-trained nets thread, this competition will allow limited use of external data, such as pre-trained nets. We will follow up with more specific rules and process on Monday, but for now our main considerations are 1) licensing (contestants and Yelp are freely able to use the data), 2) fairness (everyone has equal access to the data), 3) esstentialness (we want to see contestants' ingenuity, not scripting around an entire pre-trained model). Please expect the approval process to take a couple of business days while we learn about and discuss requests.", "103744": "Are 'caffe' and the 'pretrained BVLC reference models under the BSD 2-Clause license' fair game?\r\nhttps://github.com/BVLC/caffe/", "103915": "rbroberg You are free to use the Caffe models. Thanks for posting the request.", "107757": "Hi!\r\n\r\nCan I use this VGG16 pretrained model for Keras?\r\n\r\nhttps://gist.github.com/baraldilorenzo/07d7802847aaad0a35d3\r\n\r\nKindly regards, Dmitry", "107786": "Sure. Thanks for posting the request.", "107854": "can we use noLearn ?", "107856": "https://github.com/dnouri/nolearn", "107861": "1: Can we use \"ilsvrc12 image mean file\" for caffe models? (we can download with https://github.com/BVLC/caffe/blob/master/data/ilsvrc12/get_ilsvrc_aux.sh)\r\n\r\n2:Can we use \"channel-wise mean value of ilsvrc12 dataset\"? (Values of 18~20th-rows of https://github.com/BVLC/caffe/blob/master/models/bvlc_reference_caffenet/train_val.prototxt)\r\n\r\n3:You said we can use VGG16 model but \"https://gist.github.com/ksimonyan/211839e770f7b538e2d8#file-readme-md\" is saying \"non-commercial use only\", is it OK to use VGG16 model in this competition?", "107863": "ki thanks for bringing this up. Seems like using VGG16 is against the rule since Yelp has no right to use it due to the non-commercial license.\r\n\r\nnolearn and the caffe data is OK to use.", "107950": "Can we use theano, torch, tensor?", "107951": "", "108368": "You are free to use open-sourced packages that allow commercial usage. theano, torch, tensorflow all belongs to this category", "109071": "can we use: https://github.com/soumith/inception.torch", "109296": "Yes you can use the inception net", "110204": "OK, so I'm a little confused. Here http://www.robots.ox.ac.uk/~vgg/research/very_deep/ there is information that VGG16 and VGG19 are under CC4 with commercial usage", "110207": "But the pages\r\nhttps://gist.github.com/ksimonyan/3785162f95cd2d5fee77#file-readme-md\r\nhttps://gist.github.com/ksimonyan/211839e770f7b538e2d8#file-readme-md\r\n\r\nSuggests that they are non-commercial. Seems like it is not very consistent here.\r\n\r\nIf you are very interested in trying these VGG models please confirm with the original authors what the actual license they are using and have them update the pages. Otherwise please avoid using them for the current contest. Thanks!", "110859": "Are all Caffe models in the model zoo allowed? Are we allowed to use prebuilt or pre-trained torch models too?", "110863": "Not necessarily. The model needs to be free for commercial usage, which does not apply to certain licenses (e.g. GPL). While most models should be OK, if you are interested a certain model, please post it here and we will check the license.", "112546": "[quote=Fang-Chieh Chou;110863]\r\n\r\nNot necessarily. The model needs to be free for commercial usage, which does not apply to certain licenses (e.g. GPL). While most models should be OK, if you are interested a certain model, please post it here and we will check the license.\r\n\r\n[/quote]\r\n\r\n\r\nI am sorry, what commercial usage are you talking about? This is a Recruiting Competition, and two other admins confirmed this is about hiring and job skills only!\r\n\r\n\r\n[quote]RECRUITING COMPETITION SUBMISSION LICENSE GRANT\r\n\r\nIf the Competition is classed above as a Recruiting Competition, Participant agrees to grant and hereby grants to Competition Sponsor a perpetual, irrevocable, worldwide, royalty-free, transferable, and sublicenseable right to use, review, reproduce, and internally distribute the Submission (including any and all submitted source code) in connection with evaluating Participant's suitability for employment, provided that the Submission licensed as set forth above will **not be used for Competition Host’s commercial purposes** except as may otherwise be set forth in this Agreement.[/quote]\r\n\r\n[quote=ody;110344]\r\nJohn, we are not planning to use the winning solution commercially; this is mostly a hiring exercise, a chance to share an interesting/less-common type of machine learning problem (and dataset) with the Kaggle community, and for academic interest. [/quote]\r\n\r\n[quote=Jim B;108328]\r\n This contest is a chance to let the results speak for themselves -- if you can create great algorithms, then it doesn't matter what you have on your resume, we want to talk to you.\r\n\r\nReleasing features at Yelp is about more than writing an algorithm alone with supplied data and one metric, so we need to test for the other abilities as well. That means we can't guarantee a job offer.\r\n[/quote]", "112553": "rakhlin Thanks for asking, but the current rule also states that\r\n\r\n OPEN-SOURCE CODE\r\n \r\n A Submission will be ineligible to win a prize if it was developed using code containing or depending on software licensed under an open source license:\r\n \r\n * other than an Open Source Initiative-approved license (see