{ "id": 106567, "title": "Trying to ensemble on 5 folds using StratifiedKFold", "url": "/competitions/aptos2019-blindness-detection/discussion/106567", "author_name": "", "post_date": "2019-08-30T01:23:33.108589900Z", "votes": null, "comment_count": 4, "views": 0, "content": "

This is what I have setup for the CV.\n\"\"\nBut I keep getting this error: \n\"\"

\n\n

This is what I set x and y as : \n\"\"

\n\n

Do I just drop values from x so that they match y?

", "messages": [ { "id": "612750", "postDate": "08/30/2019 01:23:33", "content": "

This is what I have setup for the CV.\n\"\"\nBut I keep getting this error: \n\"\"

\n\n

This is what I set x and y as : \n\"\"

\n\n

Do I just drop values from x so that they match y?

", "rawMarkdown": "This is what I have setup for the CV.\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2448809%2F90dbb276b37ab0e80fb58cf4666096ea%2Fcode.PNG?generation=1567127919738295&alt=media)\nBut I keep getting this error: \n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2448809%2F537e25350f69d1767390100874cbbbc8%2F2.PNG?generation=1567128003875881&alt=media)\n\nThis is what I set x and y as : \n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2448809%2F914283c93b1dff95b2cdbddc54aa82bd%2F3.PNG?generation=1567128120385008&alt=media)\n\nDo I just drop values from x so that they match y?", "votes": null }, { "id": "612909", "postDate": "08/30/2019 04:47:30", "content": "

for fold_, (train_, val_) in enumerate(skf.split(train_csv, train_csv['diagnosis'].values)):\n df_train = train_csv.loc[train_]\n df_val = train_csv.loc[val_]\n\ntry this code may help you slove this error

", "rawMarkdown": "`for fold_, (train_, val_) in enumerate(skf.split(train_csv, train_csv['diagnosis'].values)):\n df_train = train_csv.loc[train_]\n df_val = train_csv.loc[val_]\n`\ntry this code may help you slove this error", "votes": null }, { "id": "613430", "postDate": "08/30/2019 13:52:41", "content": "

thank you I will try that

", "rawMarkdown": "thank you I will try that", "votes": null }, { "id": "615466", "postDate": "09/02/2019 01:36:12", "content": "

I have got it to work successfully but I dont see any improvement in my LB score. I dont know if I am doing it correctly. This is what I have setup \"\"

", "rawMarkdown": "I have got it to work successfully but I dont see any improvement in my LB score. I dont know if I am doing it correctly. This is what I have setup ![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2448809%2Fa5a3d6217c4bdd05894fecece41eed34%2FCapture.PNG?generation=1567388237537014&alt=media)", "votes": null }, { "id": "615479", "postDate": "09/02/2019 02:15:19", "content": "

This is trick question, I also do some test like TTA, K-Fold; the LB did not improve, sometimes I even get low score; I think the question is How your model generalize

", "rawMarkdown": "This is trick question, I also do some test like TTA, K-Fold; the LB did not improve, sometimes I even get low score; I think the question is How your model generalize", "votes": null } ], "comments": [ { "id": 612909, "author_name": "oceanwong", "author_url": "", "post_date": "08/30/2019 04:47:30", "content": "

for fold_, (train_, val_) in enumerate(skf.split(train_csv, train_csv['diagnosis'].values)):\n df_train = train_csv.loc[train_]\n df_val = train_csv.loc[val_]\n\ntry this code may help you slove this error

", "votes": null, "replies": [ { "id": 613430, "author_name": "krsoju", "author_url": "", "post_date": "08/30/2019 13:52:41", "content": "

thank you I will try that

", "votes": null, "replies": [] }, { "id": 615466, "author_name": "krsoju", "author_url": "", "post_date": "09/02/2019 01:36:12", "content": "

I have got it to work successfully but I dont see any improvement in my LB score. I dont know if I am doing it correctly. This is what I have setup \"\"

", "votes": null, "replies": [] }, { "id": 615479, "author_name": "oceanwong", "author_url": "", "post_date": "09/02/2019 02:15:19", "content": "

This is trick question, I also do some test like TTA, K-Fold; the LB did not improve, sometimes I even get low score; I think the question is How your model generalize

", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "612750": "This is what I have setup for the CV.\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2448809%2F90dbb276b37ab0e80fb58cf4666096ea%2Fcode.PNG?generation=1567127919738295&alt=media)\nBut I keep getting this error: \n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2448809%2F537e25350f69d1767390100874cbbbc8%2F2.PNG?generation=1567128003875881&alt=media)\n\nThis is what I set x and y as : \n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2448809%2F914283c93b1dff95b2cdbddc54aa82bd%2F3.PNG?generation=1567128120385008&alt=media)\n\nDo I just drop values from x so that they match y?", "612909": "`for fold_, (train_, val_) in enumerate(skf.split(train_csv, train_csv['diagnosis'].values)):\n df_train = train_csv.loc[train_]\n df_val = train_csv.loc[val_]\n`\ntry this code may help you slove this error", "613430": "thank you I will try that", "615466": "I have got it to work successfully but I dont see any improvement in my LB score. I dont know if I am doing it correctly. This is what I have setup ![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2448809%2Fa5a3d6217c4bdd05894fecece41eed34%2FCapture.PNG?generation=1567388237537014&alt=media)", "615479": "This is trick question, I also do some test like TTA, K-Fold; the LB did not improve, sometimes I even get low score; I think the question is How your model generalize" }, "source": "meta" }