Good afternoon,
\nIn order to maintain the competitiveness of the competition, we are dropping the starter dataset down to 300 training images and 95 validation images. You can still generate hundreds of training images on the FalconCloud Scenarios:
\n", "messages": [ { "id": "3247466", "postDate": "07/12/2025 19:33:41", "content": "Good afternoon,
\nIn order to maintain the competitiveness of the competition, we are dropping the starter dataset down to 300 training images and 95 validation images. You can still generate hundreds of training images on the FalconCloud Scenarios:
\n", "rawMarkdown": "Good afternoon,\n\nIn order to maintain the competitiveness of the competition, we are dropping the starter dataset down to 300 training images and 95 validation images. You can still generate hundreds of training images on the FalconCloud Scenarios:\n\n- [Multiclass Image Scenario](https://falcon.duality.ai/secure/scenarios/edit/9e90e036-8af9-41e4-8af0-1343b8e8f467)\n- [Controlled Multiclass Image Scenario](https://falcon.duality.ai/secure/scenarios/edit/e3294c19-49d4-4f64-9ca8-8373876c2c94)", "votes": null }, { "id": "3247485", "postDate": "07/12/2025 20:54:11", "content": "I just realized this. thanks for the heads up 😅
\nwhat about the previous submissions made ?
So, we are not allowed to use the first dataset? What about images from previous competitions?
", "rawMarkdown": "So, we are not allowed to use the first dataset? What about images from previous competitions?", "votes": null }, { "id": "3247609", "postDate": "07/13/2025 04:59:20", "content": "I think it is currently disabled.
\nCan @rebekahduality can you allow us to use previous datasets as they are also synthetic
@thelastsmilodon and @younesbenalia you can continue using the original dataset. It isn't very hard to generate 1000 more images on the cloud with similar parameters, so it isn't a huge advantage, more of a time barrier.
\nIf anyone feels this is unfair, reach out to me and I can provide support. We are also available to do live office hours to support effective data creation from the scenario.
@rebekahduality I'm facing a problem when I run a scenario; I'm unable to download the data, and it shows \"Session data is being prepared…\". No matter how long I wait, I can't get the data. Refreshing the page doesn't change anything.
", "rawMarkdown": "rebekahduality I'm facing a problem when I run a scenario; I'm unable to download the data, and it shows \"Session data is being prepared...\". No matter how long I wait, I can't get the data. Refreshing the page doesn't change anything.", "votes": null }, { "id": "3248558", "postDate": "07/14/2025 20:01:27", "content": "what about ones from the previous competition ?
", "rawMarkdown": "what about ones from the previous competition ?", "votes": null }, { "id": "3248568", "postDate": "07/14/2025 20:16:34", "content": "@younesbenalia Perhaps you are using almost all the available digital twin in the scenario? Happened to me also in the last competition a few times.
\nI migitated this by dropping/excluding a few. And it worked for me .
I tried that too, I only used static twins, and it didn't work either.
\n@muhammadhaaris27083, how many iterations are you using?
@younesbenalia we will be looking into that. Wait around 10 minutes, then try again.
\nCan you send me the link to your session so I can pass it to our IT team?
\nApologies for the inconvenience!
@muhammadhaaris27083 thank you for sharing this! I will pass it along to my team as well. Apologies for the inconvenience, but glad you found a way to make it work.
", "rawMarkdown": "muhammadhaaris27083 thank you for sharing this! I will pass it along to my team as well. Apologies for the inconvenience, but glad you found a way to make it work.", "votes": null }, { "id": "3248596", "postDate": "07/14/2025 21:18:26", "content": "You can use data from the previous compeition or the other FalconCloud Scenario
", "rawMarkdown": "You can use data from the previous compeition or the other FalconCloud Scenario", "votes": null }, { "id": "3249001", "postDate": "07/15/2025 14:09:51", "content": "Same issue here. I though it was solved. But I think is about sampling more than 100 at once.
", "rawMarkdown": "Same issue here. I though it was solved. But I think is about sampling more than 100 at once.", "votes": null }, { "id": "3249010", "postDate": "07/15/2025 14:14:26", "content": "Since is allowed and FALCON is not working very well ,at least for me. Could be possible to reupload and make public that data please?
", "rawMarkdown": "Since is allowed and FALCON is not working very well ,at least for me. Could be possible to reupload and make public that data please?", "votes": null }, { "id": "3249021", "postDate": "07/15/2025 14:25:01", "content": "Hi All, Apologies to everyone, we are working on the cloud issue right now and we will post the update here once we resolve the issue.
\nThank you for being patient!
", "rawMarkdown": "Hi All, Apologies to everyone, we are working on the cloud issue right now and we will post the update here once we resolve the issue.\n\nThank you for being patient!", "votes": null }, { "id": "3249036", "postDate": "07/15/2025 14:47:55", "content": "@rebekahduality https://falcon.duality.ai/secure/sessions/edit/b93432ba-985f-4d69-93cb-030f2ecf6c9e
", "rawMarkdown": "rebekahduality https://falcon.duality.ai/secure/sessions/edit/b93432ba-985f-4d69-93cb-030f2ecf6c9e", "votes": null }, { "id": "3249062", "postDate": "07/15/2025 16:02:01", "content": "We have heard you on this, and that may happen if we are unable to fix this soon. We are dedicating some more time to addressing the issue first, however. Thank you for your patience, sorry for the frustration.
", "rawMarkdown": "We have heard you on this, and that may happen if we are unable to fix this soon. We are dedicating some more time to addressing the issue first, however. Thank you for your patience, sorry for the frustration.", "votes": null }, { "id": "3249191", "postDate": "07/16/2025 00:03:29", "content": "As promised, we have re-uploaded the 1000 image dataset here on Kaggle. It will remain up for as long as we are having these technical issues, and then go back to 200 once the image generator is functioning again.
", "rawMarkdown": "As promised, we have re-uploaded the 1000 image dataset here on Kaggle. It will remain up for as long as we are having these technical issues, and then go back to 200 once the image generator is functioning again.", "votes": null }, { "id": "3249697", "postDate": "07/17/2025 00:01:14", "content": "Hello all! Our systems have been overloaded with all of your participation in this competition, which, while a good problem to have, did have to get limited so that server didn't crash.
\nWe have implemented the following changes:
\nWe also recommend keeping your scenario from running over 30 minutes.
\nPlease keep alerting us if you find any more issues. You should be good to use all or any twins at once.
\nWe also strong encourage you to use the 3LC software to find data weaknesses and generate targeted data to address those weaknesses using the Controlled Multi-Class Data Generating Scenario
", "rawMarkdown": "Hello all! Our systems have been overloaded with all of your participation in this competition, which, while a good problem to have, did have to get limited so that server didn't crash.\n\nWe have implemented the following changes:\n- You can only run 100 iterations max\n- You cannot run more than one scenario at the same time (you can still run them back to back)\n\nWe also recommend keeping your scenario from running over 30 minutes.\n\nPlease keep alerting us if you find any more issues. You should be good to use all or any twins at once.\n\nWe also strong encourage you to use the 3LC software to find data weaknesses and generate targeted data to address those weaknesses using the [Controlled Multi-Class Data Generating Scenario](https://falcon.duality.ai/secure/scenarios/edit/e3294c19-49d4-4f64-9ca8-8373876c2c94)", "votes": null } ], "comments": [ { "id": 3247485, "author_name": "thelastsmilodon", "author_url": "", "post_date": "07/12/2025 20:54:11", "content": "I just realized this. thanks for the heads up 😅
\nwhat about the previous submissions made ?
So, we are not allowed to use the first dataset? What about images from previous competitions?
", "votes": null, "replies": [ { "id": 3247609, "author_name": "thelastsmilodon", "author_url": "", "post_date": "07/13/2025 04:59:20", "content": "I think it is currently disabled.
\nCan @rebekahduality can you allow us to use previous datasets as they are also synthetic
@thelastsmilodon and @younesbenalia you can continue using the original dataset. It isn't very hard to generate 1000 more images on the cloud with similar parameters, so it isn't a huge advantage, more of a time barrier.
\nIf anyone feels this is unfair, reach out to me and I can provide support. We are also available to do live office hours to support effective data creation from the scenario.
@rebekahduality I'm facing a problem when I run a scenario; I'm unable to download the data, and it shows \"Session data is being prepared…\". No matter how long I wait, I can't get the data. Refreshing the page doesn't change anything.
", "votes": null, "replies": [ { "id": 3248568, "author_name": "muhammadhaaris27083", "author_url": "", "post_date": "07/14/2025 20:16:34", "content": "@younesbenalia Perhaps you are using almost all the available digital twin in the scenario? Happened to me also in the last competition a few times.
\nI migitated this by dropping/excluding a few. And it worked for me .
I tried that too, I only used static twins, and it didn't work either.
\n@muhammadhaaris27083, how many iterations are you using?
@muhammadhaaris27083 thank you for sharing this! I will pass it along to my team as well. Apologies for the inconvenience, but glad you found a way to make it work.
", "votes": null, "replies": [] } ] }, { "id": 3248584, "author_name": "rebekahduality", "author_url": "", "post_date": "07/14/2025 20:57:49", "content": "@younesbenalia we will be looking into that. Wait around 10 minutes, then try again.
\nCan you send me the link to your session so I can pass it to our IT team?
\nApologies for the inconvenience!
Same issue here. I though it was solved. But I think is about sampling more than 100 at once.
", "votes": null, "replies": [] }, { "id": 3249036, "author_name": "younesbenalia", "author_url": "", "post_date": "07/15/2025 14:47:55", "content": "@rebekahduality https://falcon.duality.ai/secure/sessions/edit/b93432ba-985f-4d69-93cb-030f2ecf6c9e
", "votes": null, "replies": [] } ] } ] }, { "id": 3248558, "author_name": "thelastsmilodon", "author_url": "", "post_date": "07/14/2025 20:01:27", "content": "what about ones from the previous competition ?
", "votes": null, "replies": [ { "id": 3248596, "author_name": "rebekahduality", "author_url": "", "post_date": "07/14/2025 21:18:26", "content": "You can use data from the previous compeition or the other FalconCloud Scenario
", "votes": null, "replies": [] } ] }, { "id": 3249010, "author_name": "sacuscreed", "author_url": "", "post_date": "07/15/2025 14:14:26", "content": "Since is allowed and FALCON is not working very well ,at least for me. Could be possible to reupload and make public that data please?
", "votes": null, "replies": [ { "id": 3249021, "author_name": "smmaaz", "author_url": "", "post_date": "07/15/2025 14:25:01", "content": "Hi All, Apologies to everyone, we are working on the cloud issue right now and we will post the update here once we resolve the issue.
\nThank you for being patient!
", "votes": null, "replies": [] }, { "id": 3249062, "author_name": "rebekahduality", "author_url": "", "post_date": "07/15/2025 16:02:01", "content": "We have heard you on this, and that may happen if we are unable to fix this soon. We are dedicating some more time to addressing the issue first, however. Thank you for your patience, sorry for the frustration.
", "votes": null, "replies": [] } ] } ] } ] }, { "id": 3249191, "author_name": "", "author_url": "", "post_date": "07/16/2025 00:03:29", "content": "As promised, we have re-uploaded the 1000 image dataset here on Kaggle. It will remain up for as long as we are having these technical issues, and then go back to 200 once the image generator is functioning again.
", "votes": null, "replies": [] } ] }, { "id": 3249697, "author_name": "rebekahduality", "author_url": "", "post_date": "07/17/2025 00:01:14", "content": "Hello all! Our systems have been overloaded with all of your participation in this competition, which, while a good problem to have, did have to get limited so that server didn't crash.
\nWe have implemented the following changes:
\nWe also recommend keeping your scenario from running over 30 minutes.
\nPlease keep alerting us if you find any more issues. You should be good to use all or any twins at once.
\nWe also strong encourage you to use the 3LC software to find data weaknesses and generate targeted data to address those weaknesses using the Controlled Multi-Class Data Generating Scenario
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "3247466": "Good afternoon,\n\nIn order to maintain the competitiveness of the competition, we are dropping the starter dataset down to 300 training images and 95 validation images. You can still generate hundreds of training images on the FalconCloud Scenarios:\n\n- [Multiclass Image Scenario](https://falcon.duality.ai/secure/scenarios/edit/9e90e036-8af9-41e4-8af0-1343b8e8f467)\n- [Controlled Multiclass Image Scenario](https://falcon.duality.ai/secure/scenarios/edit/e3294c19-49d4-4f64-9ca8-8373876c2c94)", "3247485": "I just realized this. thanks for the heads up 😅\nwhat about the previous submissions made ?", "3247507": "So, we are not allowed to use the first dataset? What about images from previous competitions?", "3247609": "I think it is currently disabled. \nCan @rebekahduality can you allow us to use previous datasets as they are also synthetic", "3248535": "thelastsmilodon and @younesbenalia you can continue using the original dataset. It isn't very hard to generate 1000 more images on the cloud with similar parameters, so it isn't a huge advantage, more of a time barrier.\nIf anyone feels this is unfair, reach out to me and I can provide support. We are also available to do live office hours to support effective data creation from the scenario.", "3248551": "rebekahduality I'm facing a problem when I run a scenario; I'm unable to download the data, and it shows \"Session data is being prepared...\". No matter how long I wait, I can't get the data. Refreshing the page doesn't change anything.", "3248558": "what about ones from the previous competition ?", "3248568": "younesbenalia Perhaps you are using almost all the available digital twin in the scenario? Happened to me also in the last competition a few times.\nI migitated this by dropping/excluding a few. And it worked for me .", "3248571": "I tried that too, I only used static twins, and it didn't work either.\n@muhammadhaaris27083, how many iterations are you using?", "3248584": "younesbenalia we will be looking into that. Wait around 10 minutes, then try again.\nCan you send me the link to your session so I can pass it to our IT team?\nApologies for the inconvenience!", "3248585": "muhammadhaaris27083 thank you for sharing this! I will pass it along to my team as well. Apologies for the inconvenience, but glad you found a way to make it work.", "3248596": "You can use data from the previous compeition or the other FalconCloud Scenario", "3249001": "Same issue here. I though it was solved. But I think is about sampling more than 100 at once.", "3249010": "Since is allowed and FALCON is not working very well ,at least for me. Could be possible to reupload and make public that data please?", "3249021": "Hi All, Apologies to everyone, we are working on the cloud issue right now and we will post the update here once we resolve the issue.\n\nThank you for being patient!", "3249036": "rebekahduality https://falcon.duality.ai/secure/sessions/edit/b93432ba-985f-4d69-93cb-030f2ecf6c9e", "3249062": "We have heard you on this, and that may happen if we are unable to fix this soon. We are dedicating some more time to addressing the issue first, however. Thank you for your patience, sorry for the frustration.", "3249191": "As promised, we have re-uploaded the 1000 image dataset here on Kaggle. It will remain up for as long as we are having these technical issues, and then go back to 200 once the image generator is functioning again.", "3249697": "Hello all! Our systems have been overloaded with all of your participation in this competition, which, while a good problem to have, did have to get limited so that server didn't crash.\n\nWe have implemented the following changes:\n- You can only run 100 iterations max\n- You cannot run more than one scenario at the same time (you can still run them back to back)\n\nWe also recommend keeping your scenario from running over 30 minutes.\n\nPlease keep alerting us if you find any more issues. You should be good to use all or any twins at once.\n\nWe also strong encourage you to use the 3LC software to find data weaknesses and generate targeted data to address those weaknesses using the [Controlled Multi-Class Data Generating Scenario](https://falcon.duality.ai/secure/scenarios/edit/e3294c19-49d4-4f64-9ca8-8373876c2c94)" }, "source": "meta" }