-
Hi there,
I'm looking to use xgboost as my nuisance model in my DoubleML setup and use xgboost's own mechanism for encoding categorical features (rather than having to one hot encode them myself).
…
-
### Checks
- [X] This is not a duplicate of an existing issue (please have a look through our [open issues list](https://github.com/TheOdinProject/curriculum/issues) to make sure)
- [X] I have thorou…
-
https://arxiv.org/abs/2005.14165
-
1. linear-svm
2. k-nn
3. NB
-
- [x] xgboost
- [x] ranger
- [x] rpart
- [ ] a neural net package (maybe mxnet or keras)?
- [ ] svm from e1071
- [x] naiveBayes from e1071
- [ ] kknn (knn from class and fnn do not s…
-
Please leave your github handle here so I can invite you to the organization! Thanks!
-
### Topic of the post
Share a success story from learners using Oppia
### Goal
Fundraising/donations
### Success Metrics
Clicks
### Emotion
Impact, Excitement, Inspiration, Motivation
### What…
-
It is interesting to learn by doing.
-
Apple only for now, Android coming soon
Need to decide where to put it and how to present info
Awaiting feedback from Pepper
-
### Feature Name
Addition of complex String Algorithms
### Feature Description
Addition of following complex String Algorithms:
1)Karp-Miller-Rosenberg Algorithm
2)Wu-Manber Algorithm
3)Baeza-Ya…