Jose Luis Fernández Nuevo JLFDataScience
JLFDataScience
/ Logistic_regression_pipeline.py
Last active
January 30, 2020 16:44
Logistic regression pipeline
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from sklearn.feature_selection import SelectKBest | |
| from sklearn.feature_selection import f_classif | |
| from sklearn.pipeline import Pipeline | |
| from sklearn.linear_model import LogisticRegression | |
| pipe = Pipeline([('kBest', SelectKBest(f_classif, k = 2)), | |
| ('lr', LogisticRegression())]) | |
| pipe.fit(x_train, y_train) |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from sklearn.dummy import DummyClassifier | |
| dc = DummyClassifier() | |
| dc.fit(x_train, y_train) | |
| print(u'The performance of the model is: %0.5f' % dc.score(x_test, y_test)) |
JLFDataScience
/ train_test_df.py
Created
January 30, 2020 16:39
Divide the dataset is a training set and a test set.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from sklearn.model_selection import train_test_split | |
| target = 'Target' | |
| features = list(transf_selection.columns) | |
| features.remove(target) | |
| x_train, x_test, y_train, y_test = train_test_split(transf_selection[features], transf_selection[target], random_state = 0) |
JLFDataScience
/ VIF_aplication.py
Last active
January 30, 2020 16:35
Aplication VIF function to the donation dataset
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| calculateVIF(transfusion) | |
| #generate the new dataset with the selection of variables of the VIF function | |
| transf_selection = selectDataUsingVIF(transfusion, 5) |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| transfusion['Average (c.c./months)'] = transfusion['Volume'] / transfusion['Time'] | |
| transfusion['Donations per Month'] = (transfusion['Time'] - transfusion['Recency']) / transfusion['Frequency'] | |
| transfusion['Frequent Donor'] = transfusion['Frequency'] > median(transfusion['Frequency']) | |
| plt.figure(1, figsize=(10, 10)) | |
| plt.subplot(221) | |
| plot_distribution(transfusion, 'Average (c.c./months)', 'Target') | |
| plt.subplot(222) | |
| plot_distribution(transfusion, 'Donations per Month', 'Target') | |
| plt.subplot(223) |
JLFDataScience
/ histogram_charts.py
Created
January 30, 2020 15:56
Histogram define the relasionship target whith other features
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| plt.figure(1, figsize=(10, 10)) | |
| plt.subplot(221) | |
| plot_distribution(transfusion, 'Recency', 'Target') | |
| plt.subplot(222) | |
| plot_distribution(transfusion, 'Frequency', 'Target') | |
| plt.subplot(223) | |
| plot_distribution(transfusion, 'Volume', 'Target') | |
| plt.subplot(224) | |
| plot_distribution(transfusion, 'Time', 'Target') |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from seaborn import distplot | |
| def plot_distribution(data, feature, target): | |
| min_value = floor(min(data[feature]) - 1) | |
| max_value = ceil(max(data[feature]) + 1) | |
| bins = range(int(min_value), int(max_value), int(max(1, round((max_value - min_value) / 20)))) | |
| distplot(transfusion[data[target] == 0][feature], | |
| bins = bins, |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from seaborn import pairplot | |
| pairplot(transfusion.iloc[:, 0:4], diag_kind = 'kde'); |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from seaborn import heatmap | |
| heatmap(transfusion.corr(), annot = True); |
JLFDataScience
/ import_data_donation.py
Created
January 30, 2020 11:05
Data import from url UCI repository
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import pandas as pd | |
| transfusion = pd.read_table('https://archive.ics.uci.edu/ml/machine-learning-databases/blood-transfusion/transfusion.data', sep=',') | |
| #rename columns | |
| transfusion = transfusion.rename(columns={'Recency (months)':'Recency', | |
| 'Frequency (times)':'Frequency','Monetary (c.c. blood)':'Volume', | |
| 'Time (months)':'Time', 'whether he/she donated blood in March 2007':'Target'}) | |
| transfusion.head() |