ragulpr wtte-rnn issues

ragulpr / wtte-rnn

WTTE-RNN a framework for churn and time to event prediction

MIT License

765 stars 187 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

how would one support 3 labels: win, loss, censored?

#77 HananInfiniGrow opened 1 year ago
0
preparation data for churn prediction

#76 maksimusGREEN opened 2 years ago
1
will it work for multivariate time series prediction both regression and classification

#75 Sandy4321 opened 3 years ago
0
How to label for "time to the next event" ?

#74 tks1998 closed 3 years ago
0
Numerical instability parameterization tricks

#73 kieferk opened 4 years ago
0
Update tte_util.py

#72 BraINstinct0 opened 4 years ago
0
Porting WTTE-RNN to PyTorch

#71 BraINstinct0 closed 3 years ago
2
How to use the model to predict

#70 wayfarer91 opened 4 years ago
0
Why do you use a log in the discrete weibull loss function?

#69 agoblet closed 4 years ago
0
multi variate time series : we have categorical and continues data

#68 Sandy4321 opened 4 years ago
0
References of success of the WTTE-RNN structure?

#67 SoufianeDataFan opened 4 years ago
0
Stability of loss function for left censored data

#66 sharidd closed 4 years ago
1
Revert "add alpha and beta to cumulative_hazard in conditional_excess"

#65 ragulpr opened 4 years ago
0
Weird Beta outputs

#64 Darenar opened 4 years ago
0
possible memory issue with large data

#63 baharian opened 5 years ago
0
add alpha and beta to cumulative_hazard in conditional_excess

#62 Houssem1995 closed 4 years ago
1
wtte.pipelines.data_pipeline returns wrong seq_ids

#61 michigann opened 5 years ago
0
c-index

#60 EdMwa opened 5 years ago
0
Fix Weibull loglikelihood issues

#59 michael-tsel opened 5 years ago
0
Log-likelihood for discrete Weibull distribution

#58 michael-tsel opened 5 years ago
3
Keras and Theano why?

#57 shalberd closed 4 years ago
1
Loss Function - Not the PCF?

#56 JonStevensWork opened 5 years ago
2
Is it applicable for my dataset

#55 noppGithub closed 3 years ago
1
Event with duration

#54 flip111 opened 5 years ago
0
Pre-filtering by number of events

#53 JasonTam closed 5 years ago
2
Bimodal Hazard

#52 adamhaber-atidot opened 5 years ago
4
Invalid Loss error when training on the GPU

#51 dongishan opened 5 years ago
6
Masking layer doesn't work

#50 FrancescoFrontino opened 5 years ago
1
Loss dropping to 1e-6

#49 taxus13 opened 6 years ago
7
Release wtte 1.1.2

#48 ragulpr opened 6 years ago
0
Fix for numpy+mkl

#47 taxus13 closed 6 years ago
0
Travis CI fail Python 3.6 with "KERAS_BACKEND=theano" on develop

#46 ragulpr opened 6 years ago
0
There's a Kaggle calling out for wtte-rnn weibull

#45 chris-english opened 6 years ago
3
AttributeError: module 'numpy' has no attribute 'float128'

#44 taxus13 opened 6 years ago
4
Input Data format

#43 davidgasquez closed 5 years ago
3
High-level Questions

#42 jmwoloso closed 6 years ago
6
Clip Log-Likelihood and more [WTTE 1.1 release]

#41 ragulpr closed 6 years ago
1
Saving and Loading Model

#40 nfmcclure closed 6 years ago
3
nan loss function when b approaches 0

#39 gm-spacagna opened 6 years ago
2
Should be >= K.epsilon()

#38 gm-spacagna closed 6 years ago
1
Question about preprocessing functions

#37 adam-haber opened 6 years ago
1
Different "knobs" to improve accuracy

#36 adam-haber opened 6 years ago
6
Extracting feature importance

#35 adam-haber opened 6 years ago
1
Transforming data from long format to wtte-rnn format

#34 adam-haber opened 6 years ago
5
What the proper way of handling NaN for WTTE-RNN

#33 aprotopopov opened 6 years ago
5
Question about problem setup; predicting on real-world censored data

#32 lseffer opened 6 years ago
7
Exogeneity and grouping

#31 sn3fru closed 6 years ago
1
How does wtte work seasonally?

#30 sn3fru closed 6 years ago
1
Minor adjust in shape of return

#29 sn3fru closed 6 years ago
1
Problems at replicating the CMAPPS data score

#28 as2636 opened 7 years ago
5