-
zqshu updated
8 years ago
-
Shogun needs a way to turn calibrated probability outputs of its models (SVM, RF, etc), into calibrated ones. This means that the predicted class probability actually matches the actual class probabil…
-
Can consider the following settings:
- [ ] standad classification
- [ ] data labeled by multiple annotators
- [ ] multi-label classification
For extending `calibrate_pred_probs()` to multi-label c…
-
-
Hi Jio. Looks pretty good, but I just wanted to give you a few things to work on to move forward fom here.
- the `iter_num` you made is called an epoch
- you should learn to use the pytorch datal…
-
Isotonic Regression Notebook is not rendering properly in readthedocs - large expanse of whitespace between last cells with code and the text at the end of the notebook. Is something failing to render…
-
Project Description: Elastic Net Regression Implementation and Testing
The objective of this project is to implement the Elastic Net Regression algorithm from scratch and compare its performance wi…
-
I've run HistogramBinning on several datasets and always seem to end up with an empty right-most bin on the post-calibration reliability diagram (see figure below).
![image](https://user-images.git…
-
## Ref
- CS41 [[课程主页]](https://stanfordpython.com/#lecture)
- [ ] Test-Driven Development with Python [[youtube]](https://www.obeythetestinggoat.com/book/praise.harry.html)
- 这是一本非常棒的…
-
**Describe the bug**
On my server with CentOS, I import `cuml` after `torchvision`, the interpreter reports the error below:
```
*** Error in `/home/mypath/tools/anaconda/envs/py37/bin/python': f…