Open IvanWZhang opened 3 years ago
Hi Ivan,
Thank you for reaching out and sorry for the missing file. You can download the file from here: https://www.kaggle.com/c/lish-moa/data?select=test_features.csv and then put the file path in the train_features_pl variable.
train_features_pl = pd.read_csv('YOUR DOWNLOADED FILE PATH')
It's the same as the 'test_features' variable, we duplicate it to use for pseudo-labeling, but it's exactly the same as 'test_features'.
Kind regards,
On Mon, 31 May 2021 at 14:33, IvanWZhang @.***> wrote:
One link you post in your MoA_competition/Models/MoA_TabNet.ipynb, which is following test-moa-public: https://www.kaggle.com/nicohrubec/test-public-moa now is missing. Can you send me that "test-public-moa/test_features.csv", my email is : @.*** Really need that csv file, plz.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Amiiney/MoA_competition/issues/1, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMPL2LNP2KWDZJ6XY7XBOW3TQOFZHANCNFSM45244P3A .
Hi Amiiney,
Thank you for responding in such a hurry, so much appreciation!! I'm still a little confused so I'd like to double-check I got your point. So the so-called "train_features_pl" and "test_features" actually are exactly the same CSV file, just put in different paths in order to realize the "pseudo-labeling"? That's great!
Hi Ivan,
You are welcome! Yes! That's exactly it! "test_features" and "train_features_pl" are the exact same file. test_features is used for prediction and evaluation train_features_pl is used for pseudo-labeling
One link you post in your MoA_competition/Models/MoA_TabNet.ipynb, which is following test-moa-public: https://www.kaggle.com/nicohrubec/test-public-moa now is missing. Can you send me that "test-public-moa/test_features.csv", my email is : zyftig@sina.com Really need that csv file, plz.