Thank you, sir. This is great. Your approach is really helpful because it helps me understand a full scope of concepts and techniques to apply machine learning to real life cases. Way to go!! Thank you.
Amazing videos! I just finished the Seaborn tutorials and continuing with the rest. Do you have anything on how to properly perform a groupby() function too? I am trying to create a groupby within a groupby (more basic stuff for beginners).
Hi @Data 360 YP I am loving your tutorials I guess you did not take care of dummy variable trap here I supposed it look like this raw_data_dummy = pd.get_dummies(raw_data[['Promo', 'Day_Name']], drop_first=True) new_raw_data = pd.concat([raw_data, raw_data_dummy], axis=1)
Yiannis, I have a question regarding how to deal with outliers in a data set. I recently made box plots of the features in my data set and noticed that I had instances that were considered to be outliers because they were beyond the range of the quartiles. I read online that a good way to deal with outliers is to replace the outlier values with the median value of the column. However, in one of your videos (don't remember which one,) you said that sometimes leaving the outliers in your training data set is ok because it helps to prevent you from over fitting your data. How do you usually go about handling outliers when you run across them?
Hey Brittny. First, investigate if those points are actually outliers or not. For example, if you are comparing store sales and you have 1 store that looks to be way off, go and investigate which store is that and why. Maybe it's a mistake or maybe it's a store in the most popular area. Then you decide if you want to keep or remove
if you want to represent promotion by color use this line here inside the plot palette={"No Promo": "black", "Promotion Red": "red", "Promotion Blue":'blue'}
Hello everyone! Did you liked the video? Did you find the explanations clear enough? What should I do next? Let me know in the comments below!
next topic about neural networks
please make a continuation play list for machine learning and deep learning so that it will be easy for the beginners like me
sir please continue your playlist
I chose you as my tutor Master on machine learning
Dear Sir! I've never seen such tutorials. Simply the best. Congratulation my Professor!
You ask if I have liked it... I admire it... clear explanations, lovely English... truly thanks...
Best tutorial i found in youtube so far for beginners like me to get started on linear regression.. keep up the good work
I wonder why yo have few subscribers. You deserve a million views. I will promote this channel
Truly great video, you have the best intro to machine learning I found, I wish u the best, especially more views and subs
Thank you so much 👍
Nice . Please continue making new machine learning videos
Thanks man! Will do!
Thank you for a comprehensive explanation, truly well done Yiannis!
Glad it was helpful!
Thank you, sir. This is great. Your approach is really helpful because it helps me understand a full scope of concepts and techniques to apply machine learning to real life cases. Way to go!! Thank you.
Very informative tutorial!
Thanks!
In 12:44, what are the unique values?
Nice stuff. Text was a bit small on screen, but a lesser detail;
Good stuff, thank you. So next: weighted regression, total regression and partial least squares rergression
Thanks Andrea! Good suggestions!
Amazing videos! I just finished the Seaborn tutorials and continuing with the rest. Do you have anything on how to properly perform a groupby() function too? I am trying to create a groupby within a groupby (more basic stuff for beginners).
Hey Susan, check my pandas tutorials where I go through the groupby function. Thanks!
Hi @Data 360 YP
I am loving your tutorials
I guess you did not take care of dummy variable trap here
I supposed it look like this
raw_data_dummy = pd.get_dummies(raw_data[['Promo', 'Day_Name']], drop_first=True)
new_raw_data = pd.concat([raw_data, raw_data_dummy], axis=1)
Yiannis, I have a question regarding how to deal with outliers in a data set.
I recently made box plots of the features in my data set and noticed that I had instances that were considered to be outliers because they were beyond the range of the quartiles. I read online that a good way to deal with outliers is to replace the outlier values with the median value of the column. However, in one of your videos (don't remember which one,) you said that sometimes leaving the outliers in your training data set is ok because it helps to prevent you from over fitting your data. How do you usually go about handling outliers when you run across them?
Hey Brittny. First, investigate if those points are actually outliers or not. For example, if you are comparing store sales and you have 1 store that looks to be way off, go and investigate which store is that and why. Maybe it's a mistake or maybe it's a store in the most popular area. Then you decide if you want to keep or remove
if you want to represent promotion by color use this line here inside the plot
palette={"No Promo": "black", "Promotion Red": "red", "Promotion Blue":'blue'}