gimseng
[EXE] Logistic Regression from scratch
Learning Goals
- Implement logistic regression from scratch using numpy
Exercise Statement
- The exercise will focus on implementing logistic regression model using a single sigmoid neuron.
Prerequisites
- Basic knowlegde of numpy, pandas and matplotlib
- anaconda installed to run jupyter notebook
Data source/summary:
-
Data were extracted from images that were taken from genuine and forged banknote-like specimens. For digitization, an industrial camera usually used for print inspection was used. The final images have 400x 400 pixels. Due to the object lens and distance to the investigated object gray-scale pictures with a resolution of about 660 dpi were gained. Wavelet Transform tool were used to extract features from images.
-
The dataset is available on UCL reporsitory
(Optional) Suggest/Propose Solutions
- I have implemented ipython notebook to implement it from scratch.
(Optional) Further Links/Credits to Relevant Resources:
[e.g. This exercise and solution's proposal came from a lab session from DL2020]
Great idea @iamarpan ! Please submit a PR when you are ready, but do read through the contributing guidelines and look at how some of the other projects are organized. Many thanks !
Hi @ksdkamesh99, thanks for the interest to contribute. Let me check with @iamarpan on the progress. If its almost done, then my apology, perhaps its best for @iamarpan to finish up. If its half/not much done, maybe @iamarpan and @ksdkamesh99 could collaborate? I'll let both of you comment here / discuss on discord to see how best to proceed.
@gimseng I am almost done with the project. I'll raise a pr this weekend. @ksdkamesh99 would love to collaborate with you, you can review my pr(once it's raised) and add your suggestions or improvements.
Great @iamarpan thanks for the contribution and update ! I'll let both of you take it from here. Feel free to comment/discuss here or in our discord channel. Thanks !
I have an idea of implementing a simpler logistic regression model example like cancer survial using sklearn and include some data preprocessing steps.
