Multiple Linear Regression in Python - sklearn
ฝัง
- เผยแพร่เมื่อ 5 พ.ค. 2022
- If you are a complete beginner in machine learning, please watch the video on simple linear regression from this link before and learn the basic concepts first:
• Simple Linear Regressi...
Here is the dataset used in this video:
Please feel free to check out my Data Science blog where you will find a lot of data visualization, exploratory data analysis, statistical analysis, machine learning, natural language processing, and computer vision tutorials and projects:
regenerativetoday.com/
Twitter page:
/ rashida048
Facebook Page:
regenerativetoday.com/
#linearRegression #machinelearning #datascience #dataAnalytics #python #sklearn #jupyternotebook
im glad people like you exist. I am simply not smart enough to have figured this out on my own
Very good tutorial. No nonsense and clean. Thanks
Absolutely brilliant! Your way of explaining is beyond exceptional. Thank you so much for this simplistic explanation!
from the bottom of my heart, i want to thank you for your detailed and easy to follow explanation. i dont know who you are or where you are but you have my utter respect. big thanks
I am kinda selfish type of person. Usually I donot like videos nor subscribe channels but how precise and to be the point your video was and I'm utterly impressed as this video was helpfull in clearning my concepts about MLR.
Goodluck, Best wishes. You have won a subscriber
Fantastic video.simple to understand
Very clear instruction, thanks!
Thank you for the tutorial!
I don't know who you are, but THANK you from deep heart for making this content
I would've loved for you to squeak in a Residual analysis or whatever is done after you get your R2 values from your test and train group.
Thanks for the amazing insights!
Very well explained 🎉🎉
Thanks you so much 🎉🎉🎉
excellent. very helpful. subscribed!
where can i get the dataset that you used
super helpful, appreciate it
This video is very helpful thank you so much
This video was super helpful
Thanks Dear Rashida
Data isn't my background, but these videos help me understand how to structurally get there. Is there a way to export the predicted charges into a data population for further review. Also, is there a way to adjust the scatter plot dots by a filter on one of the independent variables (i.e. any record where age is 17, make the the plot red color). Thank you!
thanks... this is awesome
Thank you, god bless
thank you for the tutorial
omg thank you queen❤
Helpful🔥
i think u can make a function to convert object name into numeric if the the data has many columns instead of writing 1 each 1 like this :
for column in df.columns:
if not pd.api.types.is_numeric_dtype(df[column]):
df[column] = df[column].astype('category')
df[column] = df[column].cat.codes
df
Thank you so much for adding this here. I used this function in some other videos as well.
how do i go about passing new values from a user?
Nice 👍
how do i plotthe fit line over the data?
Hi, I could find the data but not the code, it's not on your github?
Very good video. About the model, dont you need to check if R-square need an adjust to trust his income?
There are a few different ways to check the model prediction. R-squared error is one of them. It is common for machine learning models to use mean squared error or mean absolute error as well.
How do we access the dataset used?
thank youuuuuuuuuuuuuuuuu miss
Can you show us how to do OneHotEncoding?
Great
If I developed a model with an r-squared of 0.2. What do I do to improve the performance of the model?
Try different hyperparameters to improve the model and also different models.
Can you share the following data please
Where is the dataset???
Thank you mam for such a wonderful learning!! I want to know further how can I improve my model accuracy with train score 0.75 and test score -1.12 ??
First is trying to tune hyperparameters, and also it is normal practice to try different models to find out which model works best for the dataset. Feel free to have a look at this video where you will find a technique for hyperparameter tuning: th-cam.com/video/km71sruT9jE/w-d-xo.html
I have a Different Insight from that i used the Wine data set for that
please may i ask you why you didn't put (axis = 1) when you drop a column
Because it's the default.
x_train,x_test,y_train,y_test=train_test_split(x,y,test_size=0.2,random_state=0) it works fine but when i swapped the x_train and x_test it gives me error.
x_test,x_train,y_train,y_test=train_test_split(x,y,test_size=0.2,random_state=0) why this code gives me error. can you please explain me?
It should give you error because x_test and y_train have different sizes
@@regenerativetoday4244i dont got your point. sized are same. I wanted to know if i write x_test,x_train .... it gives me error but it i write x_train,x_test.... then it works fine.
Why my coding shows "TypeError: float() argument must be a string or a real number, not 'Timestamp'"? which one could help me to solve this problem, plz!!
You need to check the data type of all the columns. If you see any variable is coming as timestamp, that needs to be excluded. Because this tutorial didn't account for datetime datatype. There are different ways of dealing with timestamps. You will find one way of using the timestamp data in this type of models in this tutorial: th-cam.com/video/Kt9_AI12qtM/w-d-xo.html
Thank you sooooo much!!!! really helpful:)@@regenerativetoday4244
Erm, I think the method you convert the data "region" is inappropriate. U cant convert the "region" as category since it become ordinal data. I think we should convert each of the region into dummy variables then we can see the coefficient of each region.
Exactly
Fantastic video. Very simple and to the point. How can I add the regression line to the chart?
do you have the answer?
@@svea3524 let me find it later for you. I got it eventually
use plt.plot to draw regression line i.e in the format
plt.plot(X_train, reg.predict(np.column_stack((X_train))), color='blue', label='Regression Line')
Please can you send me any link for case study using python polynomial regression (or multi polynomial) with data ?
I want to practice.
Here it is: th-cam.com/video/nqNdBlA-j4w/w-d-xo.html
What if a dataset has columns with numerical values but with symbols, how to do the cleaning?
I mean comma or currency symbol, thank you
have you got any videos that calculate the mean absolute error for evaluation?
Its showing a error as "df isn't defined "
Can you please provide the link for the csv file? I'd like to practice the codes on my own as well
Here is the link to the dataset: github.com/rashida048/Machine-Learning-Tutorials-Scikit-Learn/blob/main/insurance.csv
Thanks!
@@regenerativetoday4244 thank you so much :)
Your content is amazing
Could you also upload or provide a google drive link for the data set file. It would be really helpful.
Here is the link to the dataset: github.com/rashida048/Machine-Learning-Tutorials-Scikit-Learn/blob/main/insurance.csv. I am sorry, TH-cam changed their policy for links.
@@regenerativetoday4244 Thanks a lot !!
❤
Good.. but normally we test a model with data that it hasn't seen before, and that's the test split.
On what are you typing your codes this is not vsc?Sorry i am a begginer
This is Jupyter Notebook.
Thank you so much!
what to do when data have null values?
I just added a detailed video on how to deal with null values. Here is the link: th-cam.com/video/BnfLUJkrMjs/w-d-xo.html
training and testing on the same dataset?
Why did you need to convert to category?
Because machine learning models cannot work with strings. It features and labels should be numeric
@@regenerativetoday4244
Ahh, I see. Thanks for a great video!
hey I think the formula and the logic is wrong, though implementation is right. Linear regression even though they may seem it is quite different from the just a simple linear equation. The input features what you define as X are in fact vectors. If you compile n with m training example you have a matrix rather than simple linear equation and it turns out to be a matrix multiplication.
The addition is something called bias. The W is the weight. Anyway keep up!
The bias term in machine leaning term can actually be compared with y_intercept in the linear formula and the weights as coefficients. in y = aX+c, a and X are variables that can be integers, vectors, arrays, or matrices. Same as c. The formula is the concept. I have a detailed tutorial with explanation that shows the linear regression implementation in python from scratch (no libraries), please check if you are interested: regenerativetoday.com/how-to-develop-a-linear-regression-algorithm-from-scratch-in-python/.
Can't download dataset
Here is the link: github.com/rashida048/Machine-Learning-Tutorials-Scikit-Learn/blob/main/insurance.csv
Very clear instruction, thanks!