Skip to content
Snippets Groups Projects
Commit 8a2dfa70 authored by buckl113's avatar buckl113 Committed by Colbry, Dirk
Browse files

Finished button

parent fd7e7f7d
No related branches found
No related tags found
No related merge requests found
Hide whitespace changes
Inline Side-by-side
Auto-SKLearn_AutoML/Classification.ipynb
+ 1
− 1
View file @ 8a2dfa70
%% Cell type:markdown id: tags: %% Cell type:markdown id: tags:
# Classification Using Auto-SKLearn # Classification Using Auto-SKLearn
%% Cell type:markdown id: tags: %% Cell type:markdown id: tags:
[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1Z5s0WXnjtxSi2oLxKG1ZTTcpVXqIjLyv#scrollTo=-ZrgwiL9kR_L) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/mcint170/DataTools_Tutorial_Demo/blob/main/Auto-SKLearn_AutoML/Classification.ipynb)
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
!pip install auto-sklearn !pip install auto-sklearn
``` ```
%% Cell type:markdown id: tags: %% Cell type:markdown id: tags:
If running on Google Colab: After running this cell, Click Runtime -> Restart runtime. Then you can run the following cells. If running on Google Colab: After running this cell, Click Runtime -> Restart runtime. Then you can run the following cells.
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
# imports # imports
from pprint import pprint from pprint import pprint
import sklearn.datasets import sklearn.datasets
import sklearn.metrics import sklearn.metrics
import pickle import pickle
import autosklearn.classification import autosklearn.classification
``` ```
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
# split the dataset # split the dataset
X, y = sklearn.datasets.load_breast_cancer(return_X_y=True) X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)
X_train, X_test, y_train, y_test = \ X_train, X_test, y_train, y_test = \
sklearn.model_selection.train_test_split(X, y, random_state=1) sklearn.model_selection.train_test_split(X, y, random_state=1)
``` ```
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
# Fit the classifier # Fit the classifier
automl = autosklearn.classification.AutoSklearnClassifier( automl = autosklearn.classification.AutoSklearnClassifier(
time_left_for_this_task=120, time_left_for_this_task=120,
per_run_time_limit=30, per_run_time_limit=30,
tmp_folder='/tmp/autosklearn_classification_example_tmp', tmp_folder='/tmp/autosklearn_classification_example_tmp',
) )
automl.fit(X_train, y_train, dataset_name='breast_cancer') automl.fit(X_train, y_train, dataset_name='breast_cancer')
``` ```
%% Output %% Output
AutoSklearnClassifier(per_run_time_limit=30, time_left_for_this_task=120, AutoSklearnClassifier(per_run_time_limit=30, time_left_for_this_task=120,
tmp_folder='/tmp/autosklearn_classification_example_tmp') tmp_folder='/tmp/autosklearn_classification_example_tmp')
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
# Different Models run by autosklearn # Different Models run by autosklearn
print(automl.leaderboard()) print(automl.leaderboard())
``` ```
%% Output %% Output
rank ensemble_weight type cost duration rank ensemble_weight type cost duration
model_id model_id
7 1 0.10 extra_trees 0.014184 1.502508 7 1 0.10 extra_trees 0.014184 1.502508
2 2 0.02 random_forest 0.028369 2.024807 2 2 0.02 random_forest 0.028369 2.024807
36 3 0.06 k_nearest_neighbors 0.028369 0.853534 36 3 0.06 k_nearest_neighbors 0.028369 0.853534
26 4 0.04 extra_trees 0.028369 2.240347 26 4 0.04 extra_trees 0.028369 2.240347
19 5 0.02 extra_trees 0.028369 2.791073 19 5 0.02 extra_trees 0.028369 2.791073
22 6 0.02 gradient_boosting 0.028369 1.149980 22 6 0.02 gradient_boosting 0.028369 1.149980
3 7 0.14 mlp 0.028369 1.667622 3 7 0.14 mlp 0.028369 1.667622
12 8 0.04 gradient_boosting 0.035461 1.240657 12 8 0.04 gradient_boosting 0.035461 1.240657
17 9 0.02 gradient_boosting 0.035461 1.510491 17 9 0.02 gradient_boosting 0.035461 1.510491
8 10 0.02 random_forest 0.035461 1.958862 8 10 0.02 random_forest 0.035461 1.958862
37 11 0.06 gradient_boosting 0.035461 1.585859 37 11 0.06 gradient_boosting 0.035461 1.585859
5 12 0.04 random_forest 0.035461 2.075770 5 12 0.04 random_forest 0.035461 2.075770
27 13 0.10 extra_trees 0.042553 1.910083 27 13 0.10 extra_trees 0.042553 1.910083
34 14 0.08 random_forest 0.042553 1.884860 34 14 0.08 random_forest 0.042553 1.884860
9 15 0.04 extra_trees 0.042553 1.799630 9 15 0.04 extra_trees 0.042553 1.799630
23 16 0.02 mlp 0.049645 2.405247 23 16 0.02 mlp 0.049645 2.405247
35 17 0.06 extra_trees 0.056738 1.586217 35 17 0.06 extra_trees 0.056738 1.586217
32 18 0.02 extra_trees 0.063830 1.650489 32 18 0.02 extra_trees 0.063830 1.650489
38 19 0.02 extra_trees 0.063830 2.128083 38 19 0.02 extra_trees 0.063830 2.128083
20 20 0.02 passive_aggressive 0.078014 0.774718 20 20 0.02 passive_aggressive 0.078014 0.774718
30 21 0.04 adaboost 0.078014 3.121010 30 21 0.04 adaboost 0.078014 3.121010
29 22 0.02 gaussian_nb 0.141844 1.951357 29 22 0.02 gaussian_nb 0.141844 1.951357
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
# Show the different models # Show the different models
pprint(automl.show_models(), indent=4) pprint(automl.show_models(), indent=4)
``` ```
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
# Predict the test labels # Predict the test labels
predictions = automl.predict(X_test) predictions = automl.predict(X_test)
print("Accuracy score:", sklearn.metrics.accuracy_score(y_test, predictions)) print("Accuracy score:", sklearn.metrics.accuracy_score(y_test, predictions))
``` ```
%% Output %% Output
Accuracy score: 0.9440559440559441 Accuracy score: 0.9440559440559441
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
# Export the model with the highest rank # Export the model with the highest rank
clf = automl.show_models()[7]['sklearn_classifier'] clf = automl.show_models()[7]['sklearn_classifier']
pickle.dump(clf,open('model.pickle','wb')) pickle.dump(clf,open('model.pickle','wb'))
``` ```
......
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment