Fixes#8847 #9141
Fixes#8847 #9141
Conversation
algorithms-keeper
bot
added
enhancement
There was a problem hiding this comment.
Thanks for your contribution. However, I'd really prefer that you don't just limit your fix to the univariate case, because the old implementation worked for multiple variables. Could you provide a direct-method implementation that also works for multiple variables?
| """ | ||
| import numpy as np | ||
| import requests | ||
|
|
||
|
|
||
| # Function to collect the CSGO dataset |
There was a problem hiding this comment.
| # Function to collect the CSGO dataset |
This comment is unnecessary because we already have a docstring for the function.
| # Remove the labels (headers) from the list | ||
| data.pop(0) |
There was a problem hiding this comment.
| # Remove the labels (headers) from the list | |
| data.pop(0) | |
| data.pop(0) # Remove the labels (headers) from the list |
| # Remove the labels (headers) from the list | ||
| data.pop(0) | ||
|
|
||
| # Convert data to a NumPy matrix |
There was a problem hiding this comment.
| # Convert data to a NumPy matrix |
Comment is unnecessary because this line is already clear without one
| :param theta : Feature vector (weight's for our model) | ||
| ;param return : Updated Feature's, using | ||
| curr_features - alpha_ * gradient(w.r.t. feature) | ||
| # Function to calculate Mean Absolute Error (MAE) |
There was a problem hiding this comment.
| # Function to calculate Mean Absolute Error (MAE) |
Unnecessary comment, we already have a docstring for the function
| ;param return : Updated Feature's, using | ||
| curr_features - alpha_ * gradient(w.r.t. feature) | ||
| # Function to calculate Mean Absolute Error (MAE) | ||
| def calculate_mae(predicted_y, original_y): |
There was a problem hiding this comment.
Please add type hints
| return sum(abs(y - predicted_y[i]) for i, y in enumerate(original_y)) / len( | ||
| original_y | ||
| ) |
There was a problem hiding this comment.
| return sum(abs(y - predicted_y[i]) for i, y in enumerate(original_y)) / len( | |
| original_y | |
| ) | |
| return sum(abs(y1 - y2) for y1, y2 in zip(predicted_y, original_y)) / len( | |
| original_y | |
| ) |
| ) | ||
|
|
||
|
|
||
| # Function to perform simple linear regression |
There was a problem hiding this comment.
| # Function to perform simple linear regression |
Comment is unnecessary when there's already a docstring for the function
|
|
||
|
|
||
| # Function to perform simple linear regression | ||
| def simple_solve(data_x, data_y): |
There was a problem hiding this comment.
Please add type hints
| print(f"sse is: {sse}") | ||
| print(f"Half mse is: {half_mse}") | ||
| print(f"Coefficient is: {beta_1}") | ||
| print(f"Intercept is: {beta_0}") |
There was a problem hiding this comment.
Please read the contributing guidelines. Algorithmic functions should avoid side effects, including printing. Instead, please have your function simply return the regression coefficients.
|
The title of this pull request is not self-documenting. Why force readers and reviewers to look up an issue? Closing. |
Describe your change:
Fixes #8847
Checklist: