,\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5069700%2Fd41104c7df61efae7d709ba52aec6d75%2F2025-11-06_11-02.png?generation=1762423615097610&alt=media\")
I was struggling with sending a valid submission as I followed the example:
\n\n\nYou must create a row in the submission file for each image. For images that do not contain a copy-move forgery predict the string \"authentic\". For all others submit run length encoded masks as serialized using the rle_encode function in the metric. The file should contain a header and have the following format:
\n\nid,\n,authentic\n,\n
After a few experiments, it turned out that the separator is not space but ,\n
Kernel ref: https://www.kaggle.com/code/jirkaborovec/forgery-detection-demo-submission
", "messages": [ { "id": "3312050", "postDate": "11/06/2025 10:07:36", "content": "I was struggling with sending a valid submission as I followed the example:
\n\n\nYou must create a row in the submission file for each image. For images that do not contain a copy-move forgery predict the string \"authentic\". For all others submit run length encoded masks as serialized using the rle_encode function in the metric. The file should contain a header and have the following format:
\n\nid,\n,authentic\n,\n
After a few experiments, it turned out that the separator is not space but ,\n
Kernel ref: https://www.kaggle.com/code/jirkaborovec/forgery-detection-demo-submission
", "rawMarkdown": "I was struggling with sending a valid submission as I followed the example:\n\n>You must create a row in the submission file for each image. For images that do not contain a copy-move forgery predict the string \"authentic\". For all others submit run length encoded masks as serialized using the rle_encode function in the metric. The file should contain a header and have the following format:\n>```\n>case_id,annotation\n>1,authentic\n>2,\"[123 4]\"\n>```\n\nAfter a few experiments, it turned out that the separator is not space but `,`\n\n\nKernel ref: https://www.kaggle.com/code/jirkaborovec/forgery-detection-demo-submission", "votes": null }, { "id": "3315024", "postDate": "11/10/2025 13:10:10", "content": "Hi, thanks for the comment.
\nIf you are in doubt about how to create the submission, I'd recommend using the official rle_encoding function to avoid any issues.
\nYou can check our rle_encoding implementation here: https://www.kaggle.com/code/metric/recodai-f1/
", "rawMarkdown": "Hi, thanks for the comment.\n\nIf you are in doubt about how to create the submission, I'd recommend using the official rle_encoding function to avoid any issues.\n\nYou can check our rle_encoding implementation here: [https://www.kaggle.com/code/metric/recodai-f1/](https://www.kaggle.com/code/metric/recodai-f1/)", "votes": null }, { "id": "3315040", "postDate": "11/10/2025 13:16:52", "content": "I did check that one too, but it does not have any submission.csv as an artifact to click on it…\nhttps://www.kaggle.com/code/metric/recodai-f1/output
", "rawMarkdown": "I did check that one too, but it does not have any submission.csv as an artifact to click on it...\nhttps://www.kaggle.com/code/metric/recodai-f1/output", "votes": null }, { "id": "3348978", "postDate": "11/26/2025 11:22:13", "content": "Hello there. Am I correct that the current, public leaderboard is computed only from the single test image we have at disposal, and submission.csv will only contain one entry with the rle anotated mask?
", "rawMarkdown": "Hello there. Am I correct that the current, public leaderboard is computed only from the single test image we have at disposal, and submission.csv will only contain one entry with the rle anotated mask?", "votes": null }, { "id": "3350656", "postDate": "11/27/2025 17:41:29", "content": "\n\nI did check that one too, but it does not have any submission.csv as an artifact to click on it…\n https://www.kaggle.com/code/metric/recodai-f1/output
\n
I am also suffering from the same issue.
", "rawMarkdown": "> I did check that one too, but it does not have any submission.csv as an artifact to click on it...\n> https://www.kaggle.com/code/metric/recodai-f1/output\n\nI am also suffering from the same issue.", "votes": null }, { "id": "3350983", "postDate": "11/28/2025 04:06:48", "content": "Andres - the current public leaderboard is computed on multiple test images. Currently our submission.csv only contains one entry to prevent unsupervised test leakage
", "rawMarkdown": "Andres - the current public leaderboard is computed on multiple test images. Currently our submission.csv only contains one entry to prevent unsupervised test leakage", "votes": null }, { "id": "3351008", "postDate": "11/28/2025 04:43:22", "content": "tyy\nI guess we will just read the test/ folder on the public notebook
", "rawMarkdown": "tyy\nI guess we will just read the test/ folder on the public notebook", "votes": null }, { "id": "3362644", "postDate": "12/05/2025 04:30:53", "content": "@joophillipecardenuto \nCould you give us sample code to create submission？
", "rawMarkdown": "joophillipecardenuto \nCould you give us sample code to create submission？", "votes": null }, { "id": "3362766", "postDate": "12/05/2025 08:49:37", "content": "Check the 2 last cells of https://www.kaggle.com/code/antonoof/eda-r-cnn-model
", "rawMarkdown": "Check the 2 last cells of https://www.kaggle.com/code/antonoof/eda-r-cnn-model", "votes": null }, { "id": "3362882", "postDate": "12/05/2025 14:04:49", "content": "May I ask if the correct format is \"123 4\"? It is still \"[123,4]\"
", "rawMarkdown": "May I ask if the correct format is \"123 4\"? It is still \"[123,4]\"", "votes": null }, { "id": "3364004", "postDate": "12/06/2025 08:13:01", "content": "\n\n\nthey have given us only one test image so that we can make a notebook that works on the test image folder, they judge our models on a private dataset that is not given to us to prevent leakage and the leaderboard score is based on that.
", "rawMarkdown": "they have given us only one test image so that we can make a notebook that works on the test image folder, they judge our models on a private dataset that is not given to us to prevent leakage and the leaderboard score is based on that.", "votes": null } ], "comments": [ { "id": 3315024, "author_name": "joophillipecardenuto", "author_url": "", "post_date": "11/10/2025 13:10:10", "content": "Hi, thanks for the comment.
\nIf you are in doubt about how to create the submission, I'd recommend using the official rle_encoding function to avoid any issues.
\nYou can check our rle_encoding implementation here: https://www.kaggle.com/code/metric/recodai-f1/
", "votes": null, "replies": [ { "id": 3315040, "author_name": "jirkaborovec", "author_url": "", "post_date": "11/10/2025 13:16:52", "content": "I did check that one too, but it does not have any submission.csv as an artifact to click on it…\nhttps://www.kaggle.com/code/metric/recodai-f1/output
", "votes": null, "replies": [ { "id": 3350656, "author_name": "hamzahabduljalil", "author_url": "", "post_date": "11/27/2025 17:41:29", "content": "\n\nI did check that one too, but it does not have any submission.csv as an artifact to click on it…\n https://www.kaggle.com/code/metric/recodai-f1/output
\n
I am also suffering from the same issue.
", "votes": null, "replies": [] } ] }, { "id": 3362644, "author_name": "npinpi", "author_url": "", "post_date": "12/05/2025 04:30:53", "content": "@joophillipecardenuto \nCould you give us sample code to create submission？
", "votes": null, "replies": [ { "id": 3362766, "author_name": "andreshzapke", "author_url": "", "post_date": "12/05/2025 08:49:37", "content": "Check the 2 last cells of https://www.kaggle.com/code/antonoof/eda-r-cnn-model
", "votes": null, "replies": [] } ] } ] }, { "id": 3348978, "author_name": "andreshzapke", "author_url": "", "post_date": "11/26/2025 11:22:13", "content": "Hello there. Am I correct that the current, public leaderboard is computed only from the single test image we have at disposal, and submission.csv will only contain one entry with the rle anotated mask?
", "votes": null, "replies": [ { "id": 3350983, "author_name": "returnofsputnik", "author_url": "", "post_date": "11/28/2025 04:06:48", "content": "Andres - the current public leaderboard is computed on multiple test images. Currently our submission.csv only contains one entry to prevent unsupervised test leakage
", "votes": null, "replies": [ { "id": 3351008, "author_name": "andreshzapke", "author_url": "", "post_date": "11/28/2025 04:43:22", "content": "tyy\nI guess we will just read the test/ folder on the public notebook
", "votes": null, "replies": [] } ] }, { "id": 3372631, "author_name": "mrearthworm", "author_url": "", "post_date": "12/12/2025 05:50:20", "content": "they have given us only one test image so that we can make a notebook that works on the test image folder, they judge our models on a private dataset that is not given to us to prevent leakage and the leaderboard score is based on that.
", "votes": null, "replies": [] } ] }, { "id": 3362882, "author_name": "mininume", "author_url": "", "post_date": "12/05/2025 14:04:49", "content": "May I ask if the correct format is \"123 4\"? It is still \"[123,4]\"
", "votes": null, "replies": [ { "id": 3364004, "author_name": "qifeihhh666", "author_url": "", "post_date": "12/06/2025 08:13:01", "content": "\n\n\n