Python Machine Learning with Scikit Learn - Regression || Python Machine Learning PT.3

The modelling phase of the data science lifecycle is without a doubt the most fun part, it's where you get to select a bunch of algorithms and train them to detect patterns within your data.

If you want, you can take it one step further and begin to automate some of the decisions that traditionally as a data scientist you'd have to make. In this video, we'll go through how to you can take the guess work out of selecting the right machine learning algorithm using Python Scikit Learn.

In this video you'll learn how to:
- Train Python machine learning models with Scikit Learn
- Perform automated hyperparameter tuning using GridSearchCV

Stuff mentioned in the video:

Oh, and don't forget to connect with me!

Happy coding!
Nick

P.s. Let me know how you go and drop a comment if you need a hand!

Music by Lakey Inspired

[canerkoldemir]

I just wanted to say how much I appreciated the quality of the content and the instructor's teaching style. The lessons were well-structured and easy to understand, and the instructor made complex concepts clear and engaging. I feel like I learned a lot from this video and I would highly recommend it to anyone interested in Python Machine Learning. Thanks for a great learning experience!

[jctech]

I can't understand why this channel has so few views. Maybe YouTube algorithm is playing against this channel 🤷‍♂️, but I'm completely sure it will success soon. Thanks Nicholas for such an amazing content: helpful and friendly 👌!

[amoshnin]

Great content. Definitely deserve more views and likes!

[danielbarrera]

Wow! your videos are incredibly useful, thanks for all this knowledge, I'm new in machine learning but I'm sure gonna learn a lot with your videos!!! Regards from Mexico city

[PUBUDUCG]

Hi Nick,
What are the other frameworks available for tabular data ML modeling? Anything that could be applied on excel datasheets >?

[kepenge]

Hi @nicholas,

What have you tweeked to improve the score of RF? I'm following and not seeing the GitHub code any difference from the video so, I'm not getting what have you tweeked to impove it.

Thank for your videos o_O

[jascbatalla]

Hi!! did the final modifications to improve r2 and AME scores was just adding back the inventory feature??
and can you give some hints of son feature transformation that might apply to this data set?

And Awesome videos, you use some techniques with for loops that are very cool and I didn't knew we could use!!

[AmitErandole]

What part of the preprocessing did you tweak towards the end? 
I've been following along with your notebook and I don't get the same r2 and MAE scores

[rickyu]

using a scr (support vector machine ) for regression would that only predict the hyperplane or the target values as well?

[wazclayjubuzz]

Sir, you're the best! But I have a question though, I want to ask this.
I'm not good at math.
Do I have a chance to start building machine learning models or learning machine learning?

[dramundu]

Thanks Nick, from a no DS background you have helped me a lot in getting going with DS and I'm hooked in here after completing the first series of Python for DS. However, is it possible that you share the link for the csv file used in these series?

[keidran_r]

can you please make a pycaret counterpart of that? thanks. 😊

[themcclure]

Good video, thanks!
However, I was a bit sad to see you skip over the tweaking you did to improve model performance. It doesn’t look like it’s a later part of the series either. I would like to see you release that (or a more intentional version) since dealing with a rubbish model set is something we’re all going to have to deal with often.

[angeloburias]

where is the part 1 and 2 ? Thank you!

[RudraPanchal-gzzw]

sir plz help me to solve this error what error is this
Starting training for rf.

ValueError Traceback (most recent call last)
Cell In[95], line 6
4 try:
5 print('Starting training for {}.'.format(algo))
----> 6 model.fit(X_train, y_train)
7 fit_models[algo] = model
8 print('{} has been successfully fit.'.format(algo))

File ~\anaconda3\Lib\site-packages\sklearn\base.py:1151, in _fit_context.<locals>.decorator.<locals>.wrapper(estimator, *args, **kwargs)
1144 estimator._validate_params()
1146 with config_context(
1147 skip_parameter_validation=(
1148 prefer_skip_nested_validation or global_skip_validation
1149 )
1150 ):
-> 1151 return fit_method(estimator, *args, **kwargs)

File ~\anaconda3\Lib\site-packages\sklearn\model_selection\_search.py:806, in BaseSearchCV.fit(self, X, y, groups, **fit_params)
803
804 refit_metric = self.refit
--> 806 X, y, groups = indexable(X, y, groups)
807 fit_params = _check_fit_params(X, fit_params)
809 cv_orig = check_cv(self.cv, y,

File ~\anaconda3\Lib\site-packages\sklearn\utils\validation.py:455, in indexable(*iterables)
436 """Make arrays indexable for cross-validation.
437
438 Checks consistent length, passes through None, and ensures that everything
(...)
451 sparse matrix, or dataframe) or `None`.
452 """
454 result = [_make_indexable(X) for X in iterables]
--> 455
456 return result

File ~\anaconda3\Lib\site-packages\sklearn\utils\validation.py:409, in
407 uniques = np.unique(lengths)
408 if len(uniques) > 1:
--> 409 raise ValueError(
410 "Found input variables with inconsistent numbers of samples: %r"
411 % [int(l) for l in lengths]
412 )

ValueError: Found input variables with inconsistent numbers of samples: [2721, 1167]