{ "id": 553843, "title": "Prize Money", "url": "/competitions/cidaut-ai-fake-scene-classification-2024/discussion/553843", "author_name": "", "post_date": "2024-12-28T20:28:55.169235500Z", "votes": null, "comment_count": 5, "views": 0, "content": "

I have a Question tbh….WHO IS COMPETING FOR THE PRIZE???!!!!!

\n

Like it says in the rules that its for spain but on the Leaderboard i see no one actually competing for it.
\nIve actually racked my brains for this competition and i really want to win the prize money

", "messages": [ { "id": "3082962", "postDate": "12/28/2024 20:28:55", "content": "

I have a Question tbh….WHO IS COMPETING FOR THE PRIZE???!!!!!

\n

Like it says in the rules that its for spain but on the Leaderboard i see no one actually competing for it.
\nIve actually racked my brains for this competition and i really want to win the prize money

", "rawMarkdown": "I have a Question tbh....WHO IS COMPETING FOR THE PRIZE???!!!!!\n\nLike it says in the rules that its for spain but on the Leaderboard i see no one actually competing for it.\nIve actually racked my brains for this competition and i really want to win the prize money", "votes": null }, { "id": "3093574", "postDate": "01/11/2025 06:10:21", "content": "

then regress yourself fr! 😴

", "rawMarkdown": "then regress yourself fr! 😴", "votes": null }, { "id": "3098601", "postDate": "01/16/2025 17:17:59", "content": "

nah i'd win

", "rawMarkdown": "nah i'd win", "votes": null }, { "id": "3100939", "postDate": "01/20/2025 06:01:49", "content": "

You won the 2nd place, it's amazing.
\nAnyway, could you share your solution, I look forward to it.

", "rawMarkdown": "You won the 2nd place, it's amazing. \nAnyway, could you share your solution, I look forward to it.", "votes": null }, { "id": "3103308", "postDate": "01/23/2025 10:08:49", "content": "

We implemented multiple methods and models to achieve high accuracy in our project. Initially, we utilized models such as ViT (Vision Transformer) with configurations vit_base_patch16_224 and vit_large_patch16_224, along with ResNeXt101 and a Microsoft model. These efforts alone allowed us to reach approximately 93% accuracy.

\n

To further enhance our results, we incorporated various preprocessing techniques. This step proved beneficial, boosting our accuracy to 98%. However, despite our best efforts, we could not improve beyond this threshold. The competition was prolonged, and the absence of a distinction between the public and private leaderboards complicated our assessment.

\n

In the end, I made minor adjustments to the CSV file that contained the 98% accuracy results, which yielded a score of 1. I had a doubt because some entries I believed were AI-generated turned out to be real data, and vice versa. This uncertainty contributed to my doubts about the results.

", "rawMarkdown": "We implemented multiple methods and models to achieve high accuracy in our project. Initially, we utilized models such as ViT (Vision Transformer) with configurations vit_base_patch16_224 and vit_large_patch16_224, along with ResNeXt101 and a Microsoft model. These efforts alone allowed us to reach approximately 93% accuracy.\n\nTo further enhance our results, we incorporated various preprocessing techniques. This step proved beneficial, boosting our accuracy to 98%. However, despite our best efforts, we could not improve beyond this threshold. The competition was prolonged, and the absence of a distinction between the public and private leaderboards complicated our assessment.\n\nIn the end, I made minor adjustments to the CSV file that contained the 98% accuracy results, which yielded a score of 1. I had a doubt because some entries I believed were AI-generated turned out to be real data, and vice versa. This uncertainty contributed to my doubts about the results.", "votes": null }, { "id": "3103310", "postDate": "01/23/2025 10:09:57", "content": "

There was also an approach where we actually compared the test data with the training data and compared the percentages and made our inference based on that as well

", "rawMarkdown": "There was also an approach where we actually compared the test data with the training data and compared the percentages and made our inference based on that as well", "votes": null } ], "comments": [ { "id": 3093574, "author_name": "shreyasraghav", "author_url": "", "post_date": "01/11/2025 06:10:21", "content": "

then regress yourself fr! 😴

", "votes": null, "replies": [ { "id": 3098601, "author_name": "solosid69", "author_url": "", "post_date": "01/16/2025 17:17:59", "content": "

nah i'd win

", "votes": null, "replies": [] } ] }, { "id": 3100939, "author_name": "edmondkirsch", "author_url": "", "post_date": "01/20/2025 06:01:49", "content": "

You won the 2nd place, it's amazing.
\nAnyway, could you share your solution, I look forward to it.

", "votes": null, "replies": [ { "id": 3103308, "author_name": "solosid69", "author_url": "", "post_date": "01/23/2025 10:08:49", "content": "

We implemented multiple methods and models to achieve high accuracy in our project. Initially, we utilized models such as ViT (Vision Transformer) with configurations vit_base_patch16_224 and vit_large_patch16_224, along with ResNeXt101 and a Microsoft model. These efforts alone allowed us to reach approximately 93% accuracy.

\n

To further enhance our results, we incorporated various preprocessing techniques. This step proved beneficial, boosting our accuracy to 98%. However, despite our best efforts, we could not improve beyond this threshold. The competition was prolonged, and the absence of a distinction between the public and private leaderboards complicated our assessment.

\n

In the end, I made minor adjustments to the CSV file that contained the 98% accuracy results, which yielded a score of 1. I had a doubt because some entries I believed were AI-generated turned out to be real data, and vice versa. This uncertainty contributed to my doubts about the results.

", "votes": null, "replies": [] }, { "id": 3103310, "author_name": "solosid69", "author_url": "", "post_date": "01/23/2025 10:09:57", "content": "

There was also an approach where we actually compared the test data with the training data and compared the percentages and made our inference based on that as well

", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "3082962": "I have a Question tbh....WHO IS COMPETING FOR THE PRIZE???!!!!!\n\nLike it says in the rules that its for spain but on the Leaderboard i see no one actually competing for it.\nIve actually racked my brains for this competition and i really want to win the prize money", "3093574": "then regress yourself fr! 😴", "3098601": "nah i'd win", "3100939": "You won the 2nd place, it's amazing. \nAnyway, could you share your solution, I look forward to it.", "3103308": "We implemented multiple methods and models to achieve high accuracy in our project. Initially, we utilized models such as ViT (Vision Transformer) with configurations vit_base_patch16_224 and vit_large_patch16_224, along with ResNeXt101 and a Microsoft model. These efforts alone allowed us to reach approximately 93% accuracy.\n\nTo further enhance our results, we incorporated various preprocessing techniques. This step proved beneficial, boosting our accuracy to 98%. However, despite our best efforts, we could not improve beyond this threshold. The competition was prolonged, and the absence of a distinction between the public and private leaderboards complicated our assessment.\n\nIn the end, I made minor adjustments to the CSV file that contained the 98% accuracy results, which yielded a score of 1. I had a doubt because some entries I believed were AI-generated turned out to be real data, and vice versa. This uncertainty contributed to my doubts about the results.", "3103310": "There was also an approach where we actually compared the test data with the training data and compared the percentages and made our inference based on that as well" }, "source": "meta" }