Notice: This page requires JavaScript to function properly.
Please enable JavaScript in your browser settings or update your browser.Lære Challenge: Comparing Models | Comparing Models
Classification with Python
course content

Kursinnhold

Classification with Python

Classification with Python

1. k-NN Classifier
What is ClassificationWhat is k-NNk-NN with Multiple FeaturesImplementing k-NNMulti-Class ClassificationTrain-test Split and Cross ValidationChallenge: Choosing the Best K Value
2. Logistic Regression
What is Logistic RegressionFinding the ParametersChallenge: Implementing Logistic RegressionDecision BoundaryOverfitting and RegularizationChallenge: Classifying Inseparable Data
3. Decision Tree
What is Decision TreeSplitting the NodesPreventing Overfitting in Decision TreesChallenge: Implementing a Decision Tree
4. Random Forest
What is Random ForestThe Randomness of ForestChallenge: Implementing a Random Forest
5. Comparing Models
Confusion MatrixMetricsChallenge: Comparing ModelsSummary

book
Challenge: Comparing Models

Now you'll compare the models we've covered using a single dataset — the breast cancer dataset. The target variable is the 'diagnosis' column, where 1 represents malignant and 0 represents benign cases.

You will apply GridSearchCV to each model to find the best parameters. In this task, you'll use recall as the scoring metric because minimizing false negatives is crucial. To have GridSearchCV select the best parameters based on recall, set scoring='recall'.

Oppgave

Swipe to start coding

You are given a breast cancer dataset stored as a DataFrame in the df variable.

  • Create a dictionary for GridSearchCV to iterate through [3, 5, 7, 12] values for n_neighbors and store it in the knn_params variable.
  • Create a dictionary for GridSearchCV to iterate through [0.1, 1, 10] values for C and store it in the lr_params variable.
  • Create a dictionary for GridSearchCV to iterate through [2, 4, 6, 10] values for max_depth and [1, 2, 4, 7] values for min_samples_leaf, and store it in the dt_params variable.
  • Create a dictionary for GridSearchCV to iterate through [2, 4, 6] values for max_depth and [20, 50, 100] values for n_estimators, and store it in the rf_params variable.
  • Initialize and train a GridSearchCV object for each of the model, and store the trained models in the respective variables: knn_grid, lr_grid, dt_grid, and rf_grid.

Løsning

Switch to desktopBytt til skrivebordet for virkelighetspraksisFortsett der du er med et av alternativene nedenfor
Alt var klart?

Hvordan kan vi forbedre det?

Takk for tilbakemeldingene dine!

Seksjon 5. Kapittel 3
toggle bottom row

book
Challenge: Comparing Models

Now you'll compare the models we've covered using a single dataset — the breast cancer dataset. The target variable is the 'diagnosis' column, where 1 represents malignant and 0 represents benign cases.

You will apply GridSearchCV to each model to find the best parameters. In this task, you'll use recall as the scoring metric because minimizing false negatives is crucial. To have GridSearchCV select the best parameters based on recall, set scoring='recall'.

Oppgave

Swipe to start coding

You are given a breast cancer dataset stored as a DataFrame in the df variable.

  • Create a dictionary for GridSearchCV to iterate through [3, 5, 7, 12] values for n_neighbors and store it in the knn_params variable.
  • Create a dictionary for GridSearchCV to iterate through [0.1, 1, 10] values for C and store it in the lr_params variable.
  • Create a dictionary for GridSearchCV to iterate through [2, 4, 6, 10] values for max_depth and [1, 2, 4, 7] values for min_samples_leaf, and store it in the dt_params variable.
  • Create a dictionary for GridSearchCV to iterate through [2, 4, 6] values for max_depth and [20, 50, 100] values for n_estimators, and store it in the rf_params variable.
  • Initialize and train a GridSearchCV object for each of the model, and store the trained models in the respective variables: knn_grid, lr_grid, dt_grid, and rf_grid.

Løsning

Switch to desktopBytt til skrivebordet for virkelighetspraksisFortsett der du er med et av alternativene nedenfor
Alt var klart?

Hvordan kan vi forbedre det?

Takk for tilbakemeldingene dine!

Seksjon 5. Kapittel 3
Switch to desktopBytt til skrivebordet for virkelighetspraksisFortsett der du er med et av alternativene nedenfor
Vi beklager at noe gikk galt. Hva skjedde?
some-alt