**1. What is the difference between fit, fit_transform, and transform?**

The way we use fit and predict in regression, similarly for functions that transform the data – we have fit and transform

**fit** – is used to fit parameters of the function

**transform** – transforming the data using parameters fitted with the fit function

**fit_transform** – to first fit the parameters of the function and then transform the data also

**2. How to tune a model using train, test and validation split?**

- Pick a combination of hyperparameter
- Train a model using those hyperparameters
- Find the model’s performance on the validation test
- Repeat this process for all combinations available
- Choose the model with the best validation score, and find out the final(generalized) score on the test set

**3. How to upgrade the Numpy library?**

To upgrade the numpy library, you can run:

**!pip install numpy==1.20.3 --user **in your Jupyter notebook

OR

**pip install numpy==1.20.3 –user **in Anaconda prompt

## 4. Tuning the model using Grid search is taking a long time to run. How to proceed?

Tuning a model using grid search usually takes a long time, you can try the following to get more insights.

`grid_cv = GridSearchCV(estimator=pipe, param_grid=param_grid, scoring=scorer, cv=5, n_jobs = -1, verbose = 2)`

* n_jobs = -1* can speed up the tuning process by utilizing all the CPU cores.

**will give you the number of times the model has to be fit so that you will get an idea of how much time will it take**

*verbose = 2***5.**** I am getting the same performance with both GridSearchCV and RandomizedSearchCV. How can I change this as this doesn’t look practical to me?**

Getting the same results is not incorrect, you might get the same results from both grid and random search.

However few things that can be checked in such cases are:

- If the value of
*n_iter*is greater than the possible number of combinations of hyperparameters then you will get the same results from both. - Check if you have passed the obtained value of hyperparameters while building the model.

## Leave a Reply