issues
search
kowaalczyk
/
astronomical-classification
201st place (top 19%) submission for PLAsTiCC Astronomical Classification Kaggle Competition, team MIMUW E L I T E.
https://www.kaggle.com/c/PLAsTiCC-2018
4
stars
0
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Final run
#61
kowaalczyk
closed
5 years ago
0
Add oversampling to LGBM
#60
kowaalczyk
closed
5 years ago
0
Feedforward neural net (mlp)
#59
kowaalczyk
closed
5 years ago
0
LSTM+CNN for time series classification
#58
kowaalczyk
opened
5 years ago
0
Add features from latest kernel
#57
kowaalczyk
opened
5 years ago
0
Add class 6 features
#56
kowaalczyk
opened
5 years ago
0
Add plotting of confusion matrix to our library
#55
kowaalczyk
opened
5 years ago
0
Add features
#54
kowaalczyk
opened
5 years ago
2
Build autoencoder for feature extraction (use hidden layer outputs as features to train classifier)
#53
kowaalczyk
opened
5 years ago
1
Find the best way of computing class_99 probabilities based on classifier output for other classes
#52
kowaalczyk
opened
5 years ago
0
Save feature importances to custom path in train_xgb
#51
kowaalczyk
closed
5 years ago
0
Find clusters that might allow for successful independent training of the classifiers
#50
kowaalczyk
opened
5 years ago
1
Remove irrelevant features from the currently used set
#49
kowaalczyk
closed
5 years ago
0
Add mlflow wrapper and tutorial
#48
kowaalczyk
closed
5 years ago
0
Refactor
#47
radrow
closed
5 years ago
0
Design experiments to perform over next days
#46
kowaalczyk
closed
5 years ago
0
Check for bugs in code from kernel
#45
kowaalczyk
closed
5 years ago
0
Refactor code from kernel to plasticc library
#44
kowaalczyk
closed
5 years ago
0
Setup mlflow workflow
#43
kowaalczyk
closed
5 years ago
0
Prediction scripts
#42
radrow
closed
5 years ago
0
Research which features are worth eliminating.
#41
kowaalczyk
opened
5 years ago
0
Implement MultiDataset class
#40
kowaalczyk
closed
5 years ago
1
Simple and tsfresh features matching ones extracted in the Kaggle score=1.2
#39
kowaalczyk
closed
5 years ago
1
Model training implemented to be the same as Kaggle score=1.2
#38
kowaalczyk
opened
5 years ago
2
Multi log loss from kaggle competition
#37
mkfilipiuk
closed
5 years ago
0
Generate ideas for more features and design consistent call API for feature adders.
#36
kowaalczyk
closed
5 years ago
1
Missing data imputation strategies
#35
kowaalczyk
opened
5 years ago
0
Feature/tsfresh features
#34
kowaalczyk
closed
5 years ago
0
Fixed fun args
#33
kowaalczyk
closed
5 years ago
0
Implement model-agnostic bagging ensemble.
#32
kowaalczyk
opened
5 years ago
0
Run XGB on test set and submit to Kaggle
#31
kowaalczyk
closed
5 years ago
0
LightGBM Model
#30
kowaalczyk
closed
5 years ago
0
CatBoost Model
#29
kowaalczyk
closed
5 years ago
2
"Simple" featurization leaves index columns in the middle of test data
#28
radrow
closed
5 years ago
0
Feature/simple features
#27
kowaalczyk
closed
5 years ago
2
Featurization `simple` is not compatible with new Dataset class
#26
radrow
closed
5 years ago
0
Created Dataset wrapper to ease loading csv.
#25
mkfilipiuk
closed
6 years ago
0
Script with features from the forum
#24
kowaalczyk
opened
6 years ago
0
Implement metric from Kaggle in metrics.py
#23
kowaalczyk
closed
5 years ago
0
Fix Dataset wrapper
#22
kowaalczyk
closed
5 years ago
0
Feature/simple features
#21
kowaalczyk
closed
6 years ago
1
Toplevel prediction pipeline
#20
radrow
closed
5 years ago
0
Fix requirements.txt
#19
radrow
closed
6 years ago
0
Add dataset wrapper
#18
kowaalczyk
closed
6 years ago
1
Feature/training and cross validation
#17
radrow
closed
5 years ago
3
Fixed download script
#16
radrow
closed
6 years ago
1
Combine anomaly detection with classifiers - DISCUSS AT THE NEXT MEETING
#15
kowaalczyk
opened
6 years ago
0
Tsfresh featurize script
#14
kowaalczyk
closed
5 years ago
0
Predict batch-by-batch script for XGBoost
#13
kowaalczyk
closed
5 years ago
2
Training & cross-validation script
#12
kowaalczyk
closed
5 years ago
0
Next