It is near the end of the submission. Since our public leaderboard is meaningless and many people use test data to train models, I wonder what is your strict score. What do you think the highest score should be currently. My current score is still around 0.031.
", "messages": [ { "id": 1656113, "postDate": "2022-01-19T05:01:34.710Z", "content": "It is near the end of the submission. Since our public leaderboard is meaningless and many people use test data to train models, I wonder what is your strict score. What do you think the highest score should be currently. My current score is still around 0.031.
", "rawMarkdown": "It is near the end of the submission. Since our public leaderboard is meaningless and many people use test data to train models, I wonder what is your strict score. What do you think the highest score should be currently. My current score is still around 0.031.", "votes": 12 }, { "id": 1657639, "postDate": "2022-01-20T11:02:39.567Z", "content": "You can check this discussion
", "rawMarkdown": "You can check [this](https://www.kaggle.com/c/g-research-crypto-forecasting/discussion/290265) discussion", "replies": [ { "id": 1659776, "postDate": "2022-01-22T05:39:45.990Z", "content": "@alankabisov thanks
", "rawMarkdown": "@alankabisov thanks" } ] }, { "id": 1657498, "postDate": "2022-01-20T08:28:33.490Z", "content": "My current score is 0.0418. Though evaluation on supplemental_train.csv is still runningβ¦
", "rawMarkdown": "My current score is 0.0418. Though evaluation on supplemental_train.csv is still running...", "replies": [ { "id": 1659384, "postDate": "2022-01-21T18:48:29.397Z", "content": "how is your supplemental_train test going? My score drops a lot when testing in supplemental_train. π
", "rawMarkdown": "how is your supplemental_train test going? My score drops a lot when testing in supplemental_train. π" }, { "id": 1659727, "postDate": "2022-01-22T04:33:17.250Z", "content": "Testing on the full supplemental_train.csv gave a score of 0.0403. But since this file contains approximately 4 months of data, I selected the recent 93 days from 2021-10-8 to 2022-1-10 yesterday
\nand obtained a result of 0.0369β¦β¦ But I also observed that there are plenty of missing data in this range so I changed the start date and tested my code again. It's still running.
But there were many errors in the previous supplemental_train.csv file. The host updated this file several days ago. The drop of performance of your code may be explained by these errors.
", "rawMarkdown": "But there were many errors in the previous supplemental_train.csv file. The host updated this file several days ago. The drop of performance of your code may be explained by these errors." }, { "id": 1660260, "postDate": "2022-01-22T14:35:08.690Z", "content": "My code just finished running and achieved 0.0387β¦ What's your model scoreπ
", "rawMarkdown": "My code just finished running and achieved 0.0387... What's your model scoreπ" }, { "id": 1660720, "postDate": "2022-01-22T22:27:11.347Z", "content": "the score on the whole supplemental_train is only 0.011. I think my model is overfitting the original test data.
", "rawMarkdown": "the score on the whole supplemental_train is only 0.011. I think my model is overfitting the original test data." }, { "id": 1660894, "postDate": "2022-01-23T04:34:37.597Z", "content": "Note that test data are contained in train.csv, so you'd better exclude this part of data when you tune your model. And just don't treat test data as your validation data. You should use another split.
", "rawMarkdown": "Note that test data are contained in train.csv, so you'd better exclude this part of data when you tune your model. And just don't treat test data as your validation data. You should use another split." }, { "id": 1661796, "postDate": "2022-01-23T17:47:51.810Z", "content": "I do split it though
", "rawMarkdown": "I do split it though" } ] }, { "id": 1656116, "postDate": "2022-01-19T05:13:03.060Z", "content": "0.017 π π π π π π π π
", "rawMarkdown": "0.017 π π π π π π π π", "replies": [ { "id": 1657492, "postDate": "2022-01-20T08:22:02.860Z", "content": "Are you the one participating in that Character Emotion Recognition competition held on DataFountain? You told me that we should all relabel the targetsπ. So familiar.
", "rawMarkdown": "Are you the one participating in that Character Emotion Recognition competition held on DataFountain? You told me that we should all relabel the targetsπ. So familiar." } ] }, { "id": 1657151, "postDate": "2022-01-20T00:30:36.120Z", "rawMarkdown": "", "isDeleted": true } ], "comments": [ { "id": 1657639, "author_name": "Alan Kabisov", "author_url": "", "post_date": "2022-01-20T11:02:39.567000", "content": "You can check this discussion
", "votes": 0, "replies": [ { "id": 1659776, "author_name": "Anshul Khadse", "author_url": "", "post_date": "2022-01-22T05:39:45.990000", "content": "@alankabisov thanks
", "votes": 0, "replies": [] } ] }, { "id": 1657498, "author_name": "ACMHIT", "author_url": "", "post_date": "2022-01-20T08:28:33.490000", "content": "My current score is 0.0418. Though evaluation on supplemental_train.csv is still runningβ¦
", "votes": 0, "replies": [ { "id": 1659384, "author_name": "Sihead", "author_url": "", "post_date": "2022-01-21T18:48:29.397000", "content": "how is your supplemental_train test going? My score drops a lot when testing in supplemental_train. π
", "votes": 0, "replies": [] }, { "id": 1659727, "author_name": "ACMHIT", "author_url": "", "post_date": "2022-01-22T04:33:17.250000", "content": "Testing on the full supplemental_train.csv gave a score of 0.0403. But since this file contains approximately 4 months of data, I selected the recent 93 days from 2021-10-8 to 2022-1-10 yesterday
\nand obtained a result of 0.0369β¦β¦ But I also observed that there are plenty of missing data in this range so I changed the start date and tested my code again. It's still running.
But there were many errors in the previous supplemental_train.csv file. The host updated this file several days ago. The drop of performance of your code may be explained by these errors.
", "votes": 0, "replies": [] }, { "id": 1660260, "author_name": "ACMHIT", "author_url": "", "post_date": "2022-01-22T14:35:08.690000", "content": "My code just finished running and achieved 0.0387β¦ What's your model scoreπ
", "votes": 0, "replies": [] }, { "id": 1660720, "author_name": "Sihead", "author_url": "", "post_date": "2022-01-22T22:27:11.347000", "content": "the score on the whole supplemental_train is only 0.011. I think my model is overfitting the original test data.
", "votes": 0, "replies": [] }, { "id": 1660894, "author_name": "ACMHIT", "author_url": "", "post_date": "2022-01-23T04:34:37.597000", "content": "Note that test data are contained in train.csv, so you'd better exclude this part of data when you tune your model. And just don't treat test data as your validation data. You should use another split.
", "votes": 0, "replies": [] }, { "id": 1661796, "author_name": "Sihead", "author_url": "", "post_date": "2022-01-23T17:47:51.810000", "content": "I do split it though
", "votes": 0, "replies": [] } ] }, { "id": 1656116, "author_name": "yuanzhe zhou", "author_url": "", "post_date": "2022-01-19T05:13:03.060000", "content": "0.017 π π π π π π π π
", "votes": 0, "replies": [ { "id": 1657492, "author_name": "ACMHIT", "author_url": "", "post_date": "2022-01-20T08:22:02.860000", "content": "Are you the one participating in that Character Emotion Recognition competition held on DataFountain? You told me that we should all relabel the targetsπ. So familiar.
", "votes": 0, "replies": [] } ] }, { "id": 1657151, "author_name": "", "author_url": "", "post_date": "2022-01-20T00:30:36.120000", "content": "", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "1656113": "It is near the end of the submission. Since our public leaderboard is meaningless and many people use test data to train models, I wonder what is your strict score. What do you think the highest score should be currently. My current score is still around 0.031.", "1657639": "You can check [this](https://www.kaggle.com/c/g-research-crypto-forecasting/discussion/290265) discussion", "1657498": "My current score is 0.0418. Though evaluation on supplemental_train.csv is still running...", "1656116": "0.017 π π π π π π π π", "1657151": "" } }