Grid vs Random Search Hyperparameter Tuning using Python

In this video, I will focus on two methods for hyperparameter tuning - Grid v/s Random Search and determine which one is better.

In Grid Search, we try every combination of a preset list of values of the hyper-parameters and evaluate the model for each combination. The pattern followed here is similar to the grid, where all the values are placed in the form of a matrix. Each set of parameters is taken into consideration and the accuracy is noted. Once all the combinations are evaluated, the model with the set of parameters which give the top accuracy is considered to be the best.

In Random Search, we try random combinations of the hyperparameters which are used to find the best solution for the built model. It tries random combinations of a range of values. To optimise with random search, the function is evaluated at some number of random configurations in the parameter space. The chances of finding the optimal parameter are comparatively higher in random search because of the random search pattern where the model might end up being trained on the optimised parameters without any aliasing.

If you do have any questions with what we covered in this video then feel free to ask in the comment section below & I'll do my best to answer those.

If you enjoy these tutorials & would like to support them then the easiest way is to simply like the video & give it a thumbs up & also it's a huge help to share these videos with anyone who you think would find them useful.

Please consider clicking the SUBSCRIBE button to be notified for future videos & thank you all for watching.

You can find me on:

#GridSearch #RandomSearch #HyperparameterTuning

[monocongo]

Thank you for this -- very succinct, no-nonsense, and clear.

[amanjangid]

Is random search cv works, like it makes distribution among parameters by randomly picking first 2-3 decision trees (in this eg.) and try to find how each parameter relates to accuracy, whether increasing a certain parameter increase or decrease our accuracy and based on this gives the output.

[abhinavmane]

heyy... I liked the way you tried to explain but i am a beginner and i didnt understand much... so i would like to suggest you to make it clear in a bit more basic way. Keep it up. :)

[FindMultiBagger]

Great ! Please make videos on lightGBM

[rafsunahmad]

ML engineer and data scientist in big company like google are they use libraries for their work or do their own code.

[michellelee]

would you have any demo for this for an nltk/nlp case, would grid/random be better for nlp?

[ansylpinto]

Very helpful.
Is training accuracy always supposed to be higher than test?

[kuberchaurasiya]

I tried RandomSearchCV with 100 iterations, but it gives different output every time, and they are not even close. What should I do?

[prashantchaturvedi]

Conclusion: Try both. Random search method has better chance than grid CV method.

[kumarmangalam]

Are there any other search methods like evolutionary, bayes