codingroockie

[codingroockie]

Cluster program without sklearn

I had problems with the KMeans function

[codingroockie]

from pandas import Series, DataFrame

import pandas as pd

import numpy as np

import os

import matplotlib.pylab as plt

from sklearn.model_selection import train_test_split

from sklearn.tree import DecisionTreeClassifier

from sklearn.metrics import classification_report

import sklearn.metrics

from sklearn.linear_model import LassoLarsCV

AH_data = pd.read_csv("car_usage.csv")

data_clean = AH_data.dropna()

recode1 = {1:1,2:0}

data_clean['V8'] = data_clean['V6'].map(recode1)

data_clean.dtypes

data_clean.describe()

"""

Modeling and Prediction

"""

#Split into training and testing sets

predictors = data_clean[['DATA_USAGE','V6','AMERICAN','MUSCLE','SPORT','4X4','SEDAN','HIGHLANDER','CROMED1','turbo1','transmision1','ac1','gpsl','polarized1','crashed1','TICKETS1','STATES','MECHANICS1','GAS1','VA1','BOXED1','BOUGHT','ELECTRIC','PARPRES']]

targets = data_clean.TARGET

pred_train, pred_test, tar_train, tar_test  =   train_test_split(predictors, targets, test_size=.4)

pred_train.shape

pred_test.shape

tar_train.shape

tar_test.shape

model1 = LassoLarsCV(cv=10,precompute=False).fit(pred_train,tar_train)

print(dict(zip(predictors.columns,model1.coef_)))

# plot coefficient progression

m_log_alphas = -np.log10(model1.alphas_)

ax = plt.gca()

plt.plot(m_log_alphas, model1.coef_path_.T)

plt.axvline(-np.log10(model1.alpha_), linestyle='--', color='k',

            label='alpha CV')

plt.ylabel('Regression Coefficients')

plt.xlabel('-log(alpha)')

plt.title('Regression Coefficients Progression for Lasso Paths')

# plot mean square error for each fold

m_log_alphascv = -np.log10(model1.cv_alphas_)

plt.figure()

plt.plot(m_log_alphascv, model1.mse_path_, ':')

plt.plot(m_log_alphascv, model1.mse_path_.mean(axis=-1), 'k',

         label='Average across the folds', linewidth=2)

plt.axvline(-np.log10(model1.alpha_), linestyle='--', color='k',

            label='alpha CV')

plt.legend()

plt.xlabel('-log(alpha)')

plt.ylabel('Mean squared error')

plt.title('Mean squared error on each fold')

# MSE from training and test data

from sklearn.metrics import mean_squared_error

train_error = mean_squared_error(tar_train, model1.predict(pred_train))

test_error = mean_squared_error(tar_test, model1.predict(pred_test))

print ('training data MSE')

print(train_error)

print ('test data MSE')

print(test_error)

# R-square from training and test data

rsquared_train=model1.score(pred_train,tar_train)

rsquared_test=model1.score(pred_test,tar_test)

print ('training data R-square')

print(rsquared_train)

print ('test data R-square')

print(rsquared_test)

Multiple variables tend to zero reducing the complecity of the model.

[codingroockie]

from pandas import Series, DataFrame

import pandas as pd

import numpy as np

import os

import matplotlib.pylab as plt

from sklearn.model_selection import train_test_split

from sklearn.tree import DecisionTreeClassifier

from sklearn.metrics import classification_report

import sklearn.metrics

AH_data = pd.read_csv("car_usage.csv")

data_clean = AH_data.dropna()

data_clean.dtypes

data_clean.describe()

"""

Modeling and Prediction

"""

#Split into training and testing sets

predictors = data_clean[['DATA_USAGE','V6','AMERICAN','MUSCLE','SPORT','4X4','SEDAN','HIGHLANDER','CROMED1','turbo1','transmision1','ac1','gpsl','polarized1','crashed1','TICKETS1','STATES','MECHANICS1','GAS1','VA1','BOXED1','BOUGHT','ELECTRIC','PARPRES']]

targets = data_clean.TARGET

pred_train, pred_test, tar_train, tar_test  =   train_test_split(predictors, targets, test_size=.4)

pred_train.shape

pred_test.shape

tar_train.shape

tar_test.shape

#Build model on training data

#classifier=DecisionTreeClassifier()

#classifier=classifier.fit(pred_train,tar_train)

#predictions=classifier.predict(pred_test)

#print(predictions)

#sklearn.metrics.confusion_matrix(tar_test,predictions)

#print(sklearn.metrics.accuracy_score(tar_test, predictions))

from sklearn.ensemble import RandomForestClassifier, ExtraTreesClassifier

classifier = RandomForestClassifier(n_estimators=25)

classifier = classifier.fit(pred_train,tar_train)

predictions = classifier.predict(pred_test)

predictions=classifier.predict(pred_test)

#print(predictions)

sklearn.metrics.confusion_matrix(tar_test,predictions)

print(sklearn.metrics.accuracy_score(tar_test, predictions))

model = ExtraTreesClassifier()

model.fit(pred_train,tar_train)

print(model.feature_importances_)

trees = range(25)

accuracy = np.zeros(25)

for idx in range(len(trees)):

    classifier = RandomForestClassifier(n_estimators = idx+1)

    classifier = classifier.fit(pred_train,tar_train)

    predictions = classifier.predict(pred_test)

    accuracy[idx] = sklearn.metrics.accuracy_score(tar_test,predictions)

plt.cla()

plt.plot(trees,accuracy)

plt.show()

0.8442622950819673 [0.02592401 0.01445604 0.02320977 0.0173445 0.00758759 0.00689442 0.06145175 0.05076942 0.04913949 0.11451924 0.01962096 0.0173606 0.02639438 0.05672182 0.05597494 0.05127223 0.01665533 0.07227021 0.0699886 0.06857387 0.01013534 0.05616487 0.05673264 0.05083798]

The most important influence comes from the .11 variable which is cromed1. Also, the more iteration made the better tree in the forest were created giving the best one at almost 23 iterations.

[codingroockie]

First assigment

from pandas import Series, DataFrame

import pandas as pd

import numpy as np

import os

import matplotlib.pylab as plt

from sklearn.model_selection import train_test_split

from sklearn.tree import DecisionTreeClassifier

from sklearn.metrics import classification_report

import sklearn.metrics

AH_data = pd.read_csv("tree_addhealth.csv")

data_clean = AH_data.dropna()

data_clean.dtypes

data_clean.describe()

"""

Modeling and Prediction

"""

#Split into training and testing sets

predictors = data_clean[['BIO_SEX','HISPANIC','WHITE','BLACK','NAMERICAN','ASIAN',

'age','ALCEVR1','ALCPROBS1','marever1','cocever1','inhever1','cigavail','DEP1',

'ESTEEM1','VIOL1','PASSIST','DEVIANT1','SCHCONN1','GPA1','EXPEL1','FAMCONCT','PARACTV',

'PARPRES']]

targets = data_clean.TREG1

pred_train, pred_test, tar_train, tar_test  =   train_test_split(predictors, targets, test_size=.4)

pred_train.shape

pred_test.shape

tar_train.shape

tar_test.shape

#Build model on training data

classifier=DecisionTreeClassifier()

classifier=classifier.fit(pred_train,tar_train)

predictions=classifier.predict(pred_test)

#print(predictions)

sklearn.metrics.confusion_matrix(tar_test,predictions)

print(sklearn.metrics.accuracy_score(tar_test, predictions))