Model Persistance

# coding: utf-8

# In[1]:

from sklearn.datasets import load_iris
iris_dataset = load_iris()


# In[2]:

iris_dataset.feature_names


# In[3]:

iris_dataset.target


# In[4]:

X_feature = iris_dataset.data
Y_target = iris_dataset.target


# In[5]:

X_new = [[3,5,4,1],[5,3,4,2]]


# In[6]:

from sklearn.linear_model import LogisticRegression
logreg = LogisticRegression()


# In[7]:

logreg.fit(X_feature,Y_target)


# In[8]:

logreg.predict(X_new)


# In[9]:

import pickle as pkl


# In[10]:

persist_model = pkl.dumps(logreg)
persist_model


# In[11]:

from sklearn.externals import joblib
joblib.dump(logreg,'regresfilename.pkl')


# In[12]:

new_logreg_estimator = joblib.load('regresfilename.pkl')


# In[13]:

new_logreg_estimator


# In[14]:

new_logreg_estimator.predict(X_new)