QEDan quora-insincere issues

QEDan / quora-insincere

Private team repo for Quora Insincere Questions Kaggle competition

0 stars 0 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

18 handle unknown words

#24 QEDan closed 5 years ago
0
2nd model with cnn instead of bilstm

#23 mkierans opened 5 years ago
0
add small trainable embedding layer for words

#22 mkierans opened 5 years ago
0
Improve configs for new model types

#21 QEDan opened 5 years ago
0
Restore saving of wrongest predictions

#20 QEDan opened 5 years ago
0
Code cleaned up. Kernels are submittable (but do they finish?)

#19 QEDan closed 5 years ago
0
Improve word embedding representation of "unknown" words

#18 mkierans opened 5 years ago
0
Data type error in callback

#17 QEDan opened 5 years ago
0
normalization of characters

#16 mkierans opened 5 years ago
0
additional features at character level

#15 mkierans opened 5 years ago
0
Changes to save config.txt file with configurations used.

#14 QEDan closed 5 years ago
0
8 model search

#13 QEDan closed 5 years ago
0
11 hyperparameter search

#12 QEDan closed 5 years ago
0
Hyperparameter search using kernels

#11 QEDan closed 5 years ago
2
7 configuration file

#10 QEDan closed 5 years ago
0
CNN for letter features of each word

#9 QEDan opened 5 years ago
0
Implement a search through modeling decisions

#8 QEDan closed 5 years ago
0
Create a configuration file

#7 QEDan closed 5 years ago
3
Pseudolabeling

#6 QEDan opened 5 years ago
1
Anomaly detection trained on sincere questions

#5 QEDan opened 5 years ago
1
Combine raw inputs with preprocessed inputs

#4 QEDan opened 5 years ago
1
Embedding-specific preprocessing steps

#3 QEDan opened 5 years ago
0
Handling of unknown words

#2 QEDan opened 5 years ago
6
Learnable loss function parameters

#1 QEDan opened 5 years ago
1