Can't believe this is free! it is much well explained comparing to what my lectures and tutor's did! Definitely recommended and Subscribed! Thank you so much!
you literally saved me , i have a project on cerrvical cancer prediction ML model and couldn't understand how to implement in code and how things come to be , you answered all the whys . Thank u soo much for this helpful video🤗
Hi. I am studying this at the moment and your explanation is superb. You include what is relevant and what is useful without unnecessary deviation into obscurities or irrelevancies. Your explanation is perfect, Misra. Thankyou.
Is this actually a clustering algorithm only? How does the algorithm know what we are looking for as a target prediction, Misra? How is it possible that 'target' is a column of the 'data' but not included in the dataframe which again is based on 'data' from sklearn library?
Quick question! 😊 Does the train_test_split function automatically remove the target variable from X_test, or should it be removed manually? I followed along with your video and encountered something interesting. When I didn't specify max_depth and ran the model, I got an accuracy score of 100%. I'm a bit confused and wondering whether it's related to the target variable being present in the test set. Any insights or explanations would be greatly appreciated!
Hey! Such an informative video. I just want to learn one thing, that is how do I enter a new list (consisting of only the features) and get my output as whether it is malignant or benign? Thanks a lot.
Thank you for the explanation. The feature importance plot depicts overall how important each feature is in distinguishing the two classes. Can we plot feature importance plot per class, one for malignant class and other for benign that shows feature importance score w.r.t each class, rather than whole?
Hey Ali, you are very welcome! That should be possible. I don't know off the top of my head how to do that but scikit-learn documentation should have this information.
Decision trees use supervised learning right? I don't understand at which point we tell the algorithm which is the correct data and which isn't (is the dataset already labelled)? Wouldn't we need to give the data and say this data = cancerous and give the other data and say this data = benign etc
Hey Ricardo, I'm not hosting the code anywhere, sorry :/ But everything I use in this video can easily be found on the scikit-learn documentationscikit-learn.org/stable/modules/tree.html
Hello guys, I’m not a student but have a question that I was hoping someone could help me with. Is there a minimum amount of data required per variable your testing when proceeding with this form of machine learning? Any guidance would be much appreciated
Hello, can you please help me. I'm using this decision tree model as a recommender system but my model can only recommend only one output. How can i recommend multiple outputs using only one sample data?
@@misraturp actually, we can't see the code of the sort_values function, not sure what are the parameters you provided inside. so far though great video, but would appreciate if you can help share .sort_values parameters inside at 18:03 onwards
Год назад
@@misraturp now i am stuck on the rest of the sort-value
Hey Voj, I do not have the code for this video anymore but you can find all the code on the scikit learn documentation for decision trees. scikit-learn.org/stable/modules/tree.html