Hey Giannis thanks for the video. Could you please make a video of applying sklearn models on time-series data(especially univariate) ? I think its interesting because its a bit different and there are not many examples about this on the internet.
Hi, could you please let me know how to read 60 k files of 3 TB data into jupyter notebook, please help me or refer any source, it's very urgent for my work. Could you please reply for the same. Thanks...
Great Video. As to model ideas for the future, how about a recommender model to select items based on similar attributes rather than the one everyone does to recommend movies via ratings.
This is great! I am wondering if you could use GridSearchCV instead of writing out a loop when tuning the random forest model? If you can, is the method equally as useful?
Mate, I need to say, You are a legend, seriously this is what everyone look around and can't find! So happy to have found this that I have to comment. Such a great work, deserve to be said. Congrats, I'm huge fan after that.
Hello, For the calf.best_estimator_, my output cell is not displaying all of the prams and putting …) at the end. How do i get the output to show everything?
Great video. If I have no 'new' data, so should I fit the XGBoost model on the training set (.fit(X_train,y_train)) only and then predict y with only X_test (.predict(X_test))?
Why do you take all of the data in the XG Boost clf.fit(X,Y), I have been learned to always split test, train, validation cause otherwise the model is overfitting and can't be tested??
Thank you Yiannis, great as always, I have little suggestion about previous episodes, If you can make more projects regarding sql,and joins, (project that can link excel with sql again) Also if you can make some projects with tableau, like sales insights or any projects. I didn't reach these today video yet to give my opinion 😅 Thank you
brother when im trying to load the csv file at the end of "unseen data" still its showing the exited column,i just copied the code you gave, whereas in your video there is no exited column. Whats the reason?dont your csv file has that column?
I've tried to follow this, but having problems with GraphViz. It says: AttributeError: module 'graphviz.backend' has no attribute 'ENCODING' . Does anyone have any ideas? The video is very informative, thank you
I think you made a mistake in your confusion matrix plot for the final_model when you used (classes=rf.classes_) It should be like this ==>> plot_confusion_matrix(cm_norm, classes=final_model .classes_) You should have used (final_model .classes_) not plot_confusion_matrix(cm_norm, classes=rf.classes_) My final model gave 0.79 TP when I used [classes=final_model .classes_]
Let me know what you think about this video! Which model shall we do next?
Hey Giannis thanks for the video. Could you please make a video of applying sklearn models on time-series data(especially univariate) ? I think its interesting because its a bit different and there are not many examples about this on the internet.
Hi, could you please let me know how to read 60 k files of 3 TB data into jupyter notebook, please help me or refer any source, it's very urgent for my work.
Could you please reply for the same.
Thanks...
Great Video. As to model ideas for the future, how about a recommender model to select items based on similar attributes rather than the one everyone does to recommend movies via ratings.
Can you please make a video on multilinear Regression model with deployment.
This is great! I am wondering if you could use GridSearchCV instead of writing out a loop when tuning the random forest model? If you can, is the method equally as useful?
Εύχομαι μέσα από την καρδιά μου τη βοήθεια που μας προσφέρεις να την απολαύσεις κ' με το παραπάνω...
Nase kala Konstantino!
Mate, I need to say, You are a legend, seriously this is what everyone look around and can't find!
So happy to have found this that I have to comment.
Such a great work, deserve to be said. Congrats, I'm huge fan after that.
Another excellent, well structured and informative video added to the Data science world by Yiannis who doesn't stop to impress us! Keep it rolling!!!
Thanks again!
This channel is the best. Please do more videos on other machine learning models.
Thank you for this great tutorial
Thank you so much for your videos. Please make more videos about data visualization.
Nice!! Been waiting for this!
Enjoy
Hello,
For the calf.best_estimator_, my output cell is not displaying all of the prams and putting …) at the end. How do i get the output to show everything?
Thanks Yiannis!
Keep up the good work!
Great video. If I have no 'new' data, so should I fit the XGBoost model on the training set (.fit(X_train,y_train)) only and then predict y with only X_test (.predict(X_test))?
Hello Sir, Why you are not making any videos, it was so great and helpful
very good !! your vids help a lot
Well detailed, keep it up!
That's the plan!
Why do you take all of the data in the XG Boost clf.fit(X,Y), I have been learned to always split test, train, validation cause otherwise the model is overfitting and can't be tested??
Thank you Yiannis, great as always,
I have little suggestion about previous episodes,
If you can make more projects regarding sql,and joins, (project that can link excel with sql again)
Also if you can make some projects with tableau, like sales insights or any projects.
I didn't reach these today video yet to give my opinion 😅
Thank you
Thanks for the idea!
Thank you Yiannis. Great content!
Glad you liked it!
brother when im trying to load the csv file at the end of "unseen data" still its showing the exited column,i just copied the code you gave, whereas in your video there is no exited column. Whats the reason?dont your csv file has that column?
is it okay to work with 49% of misclassified
i have same problem BTW
great content keep the good work up .Bravo
I've tried to follow this, but having problems with GraphViz. It says: AttributeError: module 'graphviz.backend' has no attribute 'ENCODING' . Does anyone have any ideas?
The video is very informative, thank you
How to give a single input for prediction
amazing work!
Thank you! Cheers!
Thank you for this great video
Glad you enjoyed it!
thank youuuu!!
Thank you.
I think you made a mistake in your confusion matrix plot for the final_model when you used (classes=rf.classes_)
It should be like this ==>> plot_confusion_matrix(cm_norm, classes=final_model .classes_)
You should have used (final_model .classes_) not plot_confusion_matrix(cm_norm, classes=rf.classes_)
My final model gave 0.79 TP when I used [classes=final_model .classes_]
Wow