this is a computer vision project that uses a custom dataset produced by me and come collegues. It uses a model with pretrained weights (YOLOv5s) in order to classify the data given. I am currently deciding future applciations of this model and it will be updated with more accuracy int eh future as I continue to diversify the image data.
this regression calculator is a work in progress
which basically takes in a csv file that has labelled columns(like kaggle)
and allows for the display of different types of regression lines
built on a flask framework using libraries
such as numpy, matplotlib.pyplot, and pandas. It uses an Adam optimizer to be able to handle the various regression line types and variance in the data.
small side project I created that generates the counter to any pokemon (Before Gen 7).
I use a large csv file to parse the data and manipulate it. I use pillow and tkinter to create the custom made GUI. I use collections to be able to quantify the information.
FAQ:
How does the counter generator logic work?
It typically sorts for resitant first so that it cant hit them, then it tries to find either a 4x or a 2x effectiveness so it can hit strong.
How does the search function work?
It primarily checks if the string fits any pokemon in the csv file. Then it checks if its found in any part of a pokemon that exists. And to predict typos it checks if it has above 75% of the same individual characters in any string in the csv file.
What side is the counter?
Due to prior confusion, Id like to clarify that the counter is on the LEFT side. the RIGHT side is the pokemon you would like to find a counter to.
Recent Fixes:
Made it recommend in a new formula, and now it reccomnds double types and not singular types only. And now it works for 95% of the Data Set.
Future Fixes: Fix why it reccomend anything for aegislash or other pokemon.
Overall, this is a small side project that I made for fun. Enjoy!