![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F252874%2F6950e9cf73d1be6ac3fddb8b3ed411f3%2FFailedtoFetcherror.jpg?generation=1582588067593727&alt=media\")
The Entry and Team Merger deadline for this competition is on March 3, 2020 at 11:59pm UTC
\n\nThis is the last day to:\n- Accept the rules to enter the competition.\n- Join or merge teams.\n- Declare external data you are using in this competition on the Official Exernal Data Disclosure Thread
\n\nThis is significantly earlier than most entry and team merger deadlines, which are often up to a week before the competition closes. So please act accordingly to ensure you do not miss out on the opportunity to join the competition.
", "messages": [ { "id": "755539", "postDate": "02/24/2020 21:35:21", "content": "The Entry and Team Merger deadline for this competition is on March 3, 2020 at 11:59pm UTC
\n\nThis is the last day to:\n- Accept the rules to enter the competition.\n- Join or merge teams.\n- Declare external data you are using in this competition on the Official Exernal Data Disclosure Thread
\n\nThis is significantly earlier than most entry and team merger deadlines, which are often up to a week before the competition closes. So please act accordingly to ensure you do not miss out on the opportunity to join the competition.
", "rawMarkdown": "The Entry and Team Merger deadline for this competition is on **March 3, 2020 at 11:59pm UTC**\n\nThis is the last day to:\n- Accept the rules to enter the competition.\n- Join or merge teams.\n- Declare external data you are using in this competition on the [Official Exernal Data Disclosure Thread](https://kaggle.com/c/deepfake-detection-challenge/discussion/121203)\n\nThis is significantly earlier than most entry and team merger deadlines, which are often up to a week before the competition closes. So please act accordingly to ensure you do not miss out on the opportunity to join the competition.", "votes": null }, { "id": "755598", "postDate": "02/24/2020 23:42:22", "content": "@juliaelliott I am not able to send any team invites/merger requests. Is there any reason for that?
\n\n
Are you trying to initiate a merger with a team that you would not be eligible to merge with (i.e. combined submissions exceeds allowable)? Please DM me with details, if you feel you should be able to merge but are not able to.
", "rawMarkdown": "Are you trying to initiate a merger with a team that you would not be eligible to merge with (i.e. combined submissions exceeds allowable)? Please DM me with details, if you feel you should be able to merge but are not able to.", "votes": null }, { "id": "755600", "postDate": "02/24/2020 23:51:50", "content": "I am unable to DM, since I have exceeded the email quota ! But I dont think that should be the case.
", "rawMarkdown": "I am unable to DM, since I have exceeded the email quota ! But I dont think that should be the case.", "votes": null }, { "id": "755817", "postDate": "02/25/2020 06:28:16", "content": "Just curious, why is this competition's deadline (for those 3 things) earlier than other comps?
", "rawMarkdown": "Just curious, why is this competition's deadline (for those 3 things) earlier than other comps?", "votes": null }, { "id": "755872", "postDate": "02/25/2020 08:04:06", "content": "emailed you the details.
", "rawMarkdown": "emailed you the details.", "votes": null }, { "id": "756480", "postDate": "02/25/2020 18:50:47", "content": "Fair question - The host preferred (to inspire fairness as best we can) that we ensure the field of competition was locked down ~a month prior to the competitions' close, with equal access to the same set of external data.
", "rawMarkdown": "Fair question - The host preferred (to inspire fairness as best we can) that we ensure the field of competition was locked down ~a month prior to the competitions' close, with equal access to the same set of external data.", "votes": null }, { "id": "758876", "postDate": "02/28/2020 09:26:24", "content": "@juliaelliott
\nI'm still confused, is it against the rule to use external data licensed for non-commercial use?
Per the rules, yes.
", "rawMarkdown": "Per the rules, yes.", "votes": null }, { "id": "759330", "postDate": "02/28/2020 22:37:19", "content": "how come ? if we declare the source and it's accessible for everyone that's not against the rules
", "rawMarkdown": "how come ? if we declare the source and it's accessible for everyone that's not against the rules", "votes": null }, { "id": "760614", "postDate": "03/01/2020 14:24:45", "content": "Hello, sorry I'm fairly new to all of this but in reference to \"declaring external data\" - I'm using a Kaggle dataset \"facenet pytorch vggface2\" along with my own locally trained model, do I need to decalre either of these within the thread? (Just wanted to clarify, thanks in advance!)
", "rawMarkdown": "Hello, sorry I'm fairly new to all of this but in reference to \"declaring external data\" - I'm using a Kaggle dataset \"facenet pytorch vggface2\" along with my own locally trained model, do I need to decalre either of these within the thread? (Just wanted to clarify, thanks in advance!)", "votes": null }, { "id": "760621", "postDate": "03/01/2020 14:34:20", "content": "Apologies for the noob question, but do we have to declare the model we are using? Or the the pre-trained weights? Do we have to share a link to the github project of the code that was used to create the model?
", "rawMarkdown": "Apologies for the noob question, but do we have to declare the model we are using? Or the the pre-trained weights? Do we have to share a link to the github project of the code that was used to create the model?", "votes": null }, { "id": "760634", "postDate": "03/01/2020 14:54:44", "content": "From the rules \"External Data. You may use data other than the Competition Data (“External Data”) to develop and test your models and Submissions. However, you will (i) ensure the External Data is available to use by all participants of the competition for purposes of the competition at no cost to the other participants and (ii) post such access to the External Data for the participants to the official competition forum prior to the Entry Deadline.\" - from this I assume that the model I am using, found on Kaggle is fine as it is readily available to all participants at no cost
", "rawMarkdown": "From the rules \"External Data. You may use data other than the Competition Data (“External Data”) to develop and test your models and Submissions. However, you will (i) ensure the External Data is available to use by all participants of the competition for purposes of the competition at no cost to the other participants and (ii) post such access to the External Data for the participants to the official competition forum prior to the Entry Deadline.\" - from this I assume that the model I am using, found on Kaggle is fine as it is readily available to all participants at no cost", "votes": null }, { "id": "760639", "postDate": "03/01/2020 15:00:26", "content": "From the above linked thread \"For clarity on the datasets that need to be declared here: You do not need to declare your own self-trained/original work models. But if you are using a dataset or pre-trained model obtained externally which is licensed with the right for you to use, then that dataset needs to be declared.
\n\nSo, for example: if you are using the imagenet dataset and efficientnet to train your model(s), you must declare imagenet and efficientnet on this thread, but you don’t need to make your trained model public.\"
\n\nSo therefore only pre trained models (as opposed to locally trained models I guess) will need to be declared
", "rawMarkdown": "From the above linked thread \"For clarity on the datasets that need to be declared here: You do not need to declare your own self-trained/original work models. But if you are using a dataset or pre-trained model obtained externally which is licensed with the right for you to use, then that dataset needs to be declared.\n\nSo, for example: if you are using the imagenet dataset and efficientnet to train your model(s), you must declare imagenet and efficientnet on this thread, but you don’t need to make your trained model public.\"\n\nSo therefore only pre trained models (as opposed to locally trained models I guess) will need to be declared", "votes": null }, { "id": "760794", "postDate": "03/01/2020 18:18:03", "content": "Do we need to declare(make public) our self-preprocessed data(i.e after cropping out face)? Thank you in advance.
", "rawMarkdown": "Do we need to declare(make public) our self-preprocessed data(i.e after cropping out face)? Thank you in advance.", "votes": null } ], "comments": [ { "id": 755598, "author_name": "skylord", "author_url": "", "post_date": "02/24/2020 23:42:22", "content": "@juliaelliott I am not able to send any team invites/merger requests. Is there any reason for that?
\n\n
Are you trying to initiate a merger with a team that you would not be eligible to merge with (i.e. combined submissions exceeds allowable)? Please DM me with details, if you feel you should be able to merge but are not able to.
", "votes": null, "replies": [] }, { "id": 755600, "author_name": "skylord", "author_url": "", "post_date": "02/24/2020 23:51:50", "content": "I am unable to DM, since I have exceeded the email quota ! But I dont think that should be the case.
", "votes": null, "replies": [] }, { "id": 755872, "author_name": "skylord", "author_url": "", "post_date": "02/25/2020 08:04:06", "content": "emailed you the details.
", "votes": null, "replies": [] } ] }, { "id": 755817, "author_name": "cdeotte", "author_url": "", "post_date": "02/25/2020 06:28:16", "content": "Just curious, why is this competition's deadline (for those 3 things) earlier than other comps?
", "votes": null, "replies": [ { "id": 756480, "author_name": "juliaelliott", "author_url": "", "post_date": "02/25/2020 18:50:47", "content": "Fair question - The host preferred (to inspire fairness as best we can) that we ensure the field of competition was locked down ~a month prior to the competitions' close, with equal access to the same set of external data.
", "votes": null, "replies": [] } ] }, { "id": 758876, "author_name": "chmhzc", "author_url": "", "post_date": "02/28/2020 09:26:24", "content": "@juliaelliott
\nI'm still confused, is it against the rule to use external data licensed for non-commercial use?
Per the rules, yes.
", "votes": null, "replies": [] }, { "id": 759330, "author_name": "basharallabadi", "author_url": "", "post_date": "02/28/2020 22:37:19", "content": "how come ? if we declare the source and it's accessible for everyone that's not against the rules
", "votes": null, "replies": [] } ] }, { "id": 760614, "author_name": "rah123123", "author_url": "", "post_date": "03/01/2020 14:24:45", "content": "Hello, sorry I'm fairly new to all of this but in reference to \"declaring external data\" - I'm using a Kaggle dataset \"facenet pytorch vggface2\" along with my own locally trained model, do I need to decalre either of these within the thread? (Just wanted to clarify, thanks in advance!)
", "votes": null, "replies": [ { "id": 760634, "author_name": "rah123123", "author_url": "", "post_date": "03/01/2020 14:54:44", "content": "From the rules \"External Data. You may use data other than the Competition Data (“External Data”) to develop and test your models and Submissions. However, you will (i) ensure the External Data is available to use by all participants of the competition for purposes of the competition at no cost to the other participants and (ii) post such access to the External Data for the participants to the official competition forum prior to the Entry Deadline.\" - from this I assume that the model I am using, found on Kaggle is fine as it is readily available to all participants at no cost
", "votes": null, "replies": [] } ] }, { "id": 760621, "author_name": "akashnandi", "author_url": "", "post_date": "03/01/2020 14:34:20", "content": "Apologies for the noob question, but do we have to declare the model we are using? Or the the pre-trained weights? Do we have to share a link to the github project of the code that was used to create the model?
", "votes": null, "replies": [ { "id": 760639, "author_name": "rah123123", "author_url": "", "post_date": "03/01/2020 15:00:26", "content": "From the above linked thread \"For clarity on the datasets that need to be declared here: You do not need to declare your own self-trained/original work models. But if you are using a dataset or pre-trained model obtained externally which is licensed with the right for you to use, then that dataset needs to be declared.
\n\nSo, for example: if you are using the imagenet dataset and efficientnet to train your model(s), you must declare imagenet and efficientnet on this thread, but you don’t need to make your trained model public.\"
\n\nSo therefore only pre trained models (as opposed to locally trained models I guess) will need to be declared
", "votes": null, "replies": [] } ] }, { "id": 760794, "author_name": "unkownhihi", "author_url": "", "post_date": "03/01/2020 18:18:03", "content": "Do we need to declare(make public) our self-preprocessed data(i.e after cropping out face)? Thank you in advance.
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "755539": "The Entry and Team Merger deadline for this competition is on **March 3, 2020 at 11:59pm UTC**\n\nThis is the last day to:\n- Accept the rules to enter the competition.\n- Join or merge teams.\n- Declare external data you are using in this competition on the [Official Exernal Data Disclosure Thread](https://kaggle.com/c/deepfake-detection-challenge/discussion/121203)\n\nThis is significantly earlier than most entry and team merger deadlines, which are often up to a week before the competition closes. So please act accordingly to ensure you do not miss out on the opportunity to join the competition.", "755598": "juliaelliott I am not able to send any team invites/merger requests. Is there any reason for that?\n\n", "755599": "Are you trying to initiate a merger with a team that you would not be eligible to merge with (i.e. combined submissions exceeds allowable)? Please DM me with details, if you feel you should be able to merge but are not able to.", "755600": "I am unable to DM, since I have exceeded the email quota ! But I dont think that should be the case.", "755817": "Just curious, why is this competition's deadline (for those 3 things) earlier than other comps?", "755872": "emailed you the details.", "756480": "Fair question - The host preferred (to inspire fairness as best we can) that we ensure the field of competition was locked down ~a month prior to the competitions' close, with equal access to the same set of external data.", "758876": "juliaelliott \nI'm still confused, is it against the rule to use external data licensed for **non-commercial** use?", "759106": "Per the rules, yes.", "759330": "how come ? if we declare the source and it's accessible for everyone that's not against the rules", "760614": "Hello, sorry I'm fairly new to all of this but in reference to \"declaring external data\" - I'm using a Kaggle dataset \"facenet pytorch vggface2\" along with my own locally trained model, do I need to decalre either of these within the thread? (Just wanted to clarify, thanks in advance!)", "760621": "Apologies for the noob question, but do we have to declare the model we are using? Or the the pre-trained weights? Do we have to share a link to the github project of the code that was used to create the model?", "760634": "From the rules \"External Data. You may use data other than the Competition Data (“External Data”) to develop and test your models and Submissions. However, you will (i) ensure the External Data is available to use by all participants of the competition for purposes of the competition at no cost to the other participants and (ii) post such access to the External Data for the participants to the official competition forum prior to the Entry Deadline.\" - from this I assume that the model I am using, found on Kaggle is fine as it is readily available to all participants at no cost", "760639": "From the above linked thread \"For clarity on the datasets that need to be declared here: You do not need to declare your own self-trained/original work models. But if you are using a dataset or pre-trained model obtained externally which is licensed with the right for you to use, then that dataset needs to be declared.\n\nSo, for example: if you are using the imagenet dataset and efficientnet to train your model(s), you must declare imagenet and efficientnet on this thread, but you don’t need to make your trained model public.\"\n\nSo therefore only pre trained models (as opposed to locally trained models I guess) will need to be declared", "760794": "Do we need to declare(make public) our self-preprocessed data(i.e after cropping out face)? Thank you in advance." }, "source": "meta" }