Can you answer some questions?
\n\nCan I use the face recognition library in my submission? Because I don't know the standard of 'automl' mentioned in the notice.
Can I use the example on the website of TensorFlow when I make a neural network? Or can I modify the 'Google Inception Model'?
I ask questions because I don't know what interferes with our award.\nI am sorry that interpretation is not smooth using translator.
", "messages": [ { "id": "720995", "postDate": "01/16/2020 23:23:18", "content": "Can you answer some questions?
\n\nCan I use the face recognition library in my submission? Because I don't know the standard of 'automl' mentioned in the notice.
Can I use the example on the website of TensorFlow when I make a neural network? Or can I modify the 'Google Inception Model'?
I ask questions because I don't know what interferes with our award.\nI am sorry that interpretation is not smooth using translator.
", "rawMarkdown": "Can you answer some questions?\n\n1. Can I use the face recognition library in my submission? Because I don't know the standard of 'automl' mentioned in the notice.\n\n2. Can I use the example on the website of TensorFlow when I make a neural network? Or can I modify the 'Google Inception Model'?\n\nI ask questions because I don't know what interferes with our award.\nI am sorry that interpretation is not smooth using translator.", "votes": null }, { "id": "721866", "postDate": "01/17/2020 19:51:12", "content": "If you can upload it to a dataset on Kaggle (and the total size is under 1 GB), you can use it in your kernel. (Plus any freely available resources you're using, such as Inception, need to be declared in the External Data thread.)
", "rawMarkdown": "If you can upload it to a dataset on Kaggle (and the total size is under 1 GB), you can use it in your kernel. (Plus any freely available resources you're using, such as Inception, need to be declared in the External Data thread.)", "votes": null }, { "id": "722055", "postDate": "01/18/2020 03:19:55", "content": "What @humananalog shared is all accurate. I’ll also reiterate that you are free to train offline and upload that trained model as external data into your Kaggle notebook submission. And any datasets you might use in that training or building your model should be able to be used by all participants and publicly declared on the external data thread. You do not need to “declare” your trained model, just the input original datasets used in your model.
\n\nIn addition, the “AutoML” rules allow you to use such tools in your model (like Google’s AutoML), as long as you can share the associated parameters.
", "rawMarkdown": "What @humananalog shared is all accurate. I’ll also reiterate that you are free to train offline and upload that trained model as external data into your Kaggle notebook submission. And any datasets you might use in that training or building your model should be able to be used by all participants and publicly declared on the external data thread. You do not need to “declare” your trained model, just the input original datasets used in your model. \n\nIn addition, the “AutoML” rules allow you to use such tools in your model (like Google’s AutoML), as long as you can share the associated parameters.", "votes": null }, { "id": "722074", "postDate": "01/18/2020 03:59:37", "content": "Can I ask what this means?
\n\n\n\n\n\"No using other Kaggle notebooks or utility scripts as inputs to your submission notebook.\"
\n
When we run the kernel, it is only one notebook and so I don't see where the other Kaggle notebooks would come in.
", "rawMarkdown": "Can I ask what this means?\n\n> \"No using other Kaggle notebooks or utility scripts as inputs to your submission notebook.\"\n\nWhen we run the kernel, it is only one notebook and so I don't see where the other Kaggle notebooks would come in.", "votes": null }, { "id": "722708", "postDate": "01/19/2020 00:36:30", "content": "Sadly in reality you can, it will work on your local Kernel, but despite many attempts it doesn't run when submitting. This competition has something to it.. I won't say it's badly organized. I think we just keep the following Mantra. Just work with what you have, and hope for the best.
", "rawMarkdown": "Sadly in reality you can, it will work on your local Kernel, but despite many attempts it doesn't run when submitting. This competition has something to it.. I won't say it's badly organized. I think we just keep the following Mantra. Just work with what you have, and hope for the best.", "votes": null } ], "comments": [ { "id": 721866, "author_name": "humananalog", "author_url": "", "post_date": "01/17/2020 19:51:12", "content": "If you can upload it to a dataset on Kaggle (and the total size is under 1 GB), you can use it in your kernel. (Plus any freely available resources you're using, such as Inception, need to be declared in the External Data thread.)
", "votes": null, "replies": [] }, { "id": 722055, "author_name": "juliaelliott", "author_url": "", "post_date": "01/18/2020 03:19:55", "content": "What @humananalog shared is all accurate. I’ll also reiterate that you are free to train offline and upload that trained model as external data into your Kaggle notebook submission. And any datasets you might use in that training or building your model should be able to be used by all participants and publicly declared on the external data thread. You do not need to “declare” your trained model, just the input original datasets used in your model.
\n\nIn addition, the “AutoML” rules allow you to use such tools in your model (like Google’s AutoML), as long as you can share the associated parameters.
", "votes": null, "replies": [ { "id": 722074, "author_name": "petewills", "author_url": "", "post_date": "01/18/2020 03:59:37", "content": "Can I ask what this means?
\n\n\n\n\n\"No using other Kaggle notebooks or utility scripts as inputs to your submission notebook.\"
\n
When we run the kernel, it is only one notebook and so I don't see where the other Kaggle notebooks would come in.
", "votes": null, "replies": [] } ] }, { "id": 722708, "author_name": "abualabed", "author_url": "", "post_date": "01/19/2020 00:36:30", "content": "Sadly in reality you can, it will work on your local Kernel, but despite many attempts it doesn't run when submitting. This competition has something to it.. I won't say it's badly organized. I think we just keep the following Mantra. Just work with what you have, and hope for the best.
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "720995": "Can you answer some questions?\n\n1. Can I use the face recognition library in my submission? Because I don't know the standard of 'automl' mentioned in the notice.\n\n2. Can I use the example on the website of TensorFlow when I make a neural network? Or can I modify the 'Google Inception Model'?\n\nI ask questions because I don't know what interferes with our award.\nI am sorry that interpretation is not smooth using translator.", "721866": "If you can upload it to a dataset on Kaggle (and the total size is under 1 GB), you can use it in your kernel. (Plus any freely available resources you're using, such as Inception, need to be declared in the External Data thread.)", "722055": "What @humananalog shared is all accurate. I’ll also reiterate that you are free to train offline and upload that trained model as external data into your Kaggle notebook submission. And any datasets you might use in that training or building your model should be able to be used by all participants and publicly declared on the external data thread. You do not need to “declare” your trained model, just the input original datasets used in your model. \n\nIn addition, the “AutoML” rules allow you to use such tools in your model (like Google’s AutoML), as long as you can share the associated parameters.", "722074": "Can I ask what this means?\n\n> \"No using other Kaggle notebooks or utility scripts as inputs to your submission notebook.\"\n\nWhen we run the kernel, it is only one notebook and so I don't see where the other Kaggle notebooks would come in.", "722708": "Sadly in reality you can, it will work on your local Kernel, but despite many attempts it doesn't run when submitting. This competition has something to it.. I won't say it's badly organized. I think we just keep the following Mantra. Just work with what you have, and hope for the best." }, "source": "meta" }