#!/usr/bin/env python
# coding: utf-8

# In[39]:


get_ipython().run_line_magic('matplotlib', 'inline')
import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

sns.set(style='ticks')


# ## Data import

# In[2]:


DATA_PATH = '../data/NCS/'


# In[13]:


teaching_feeding = pd.read_csv(DATA_PATH + 'ncs_teaching_feeding_v1.csv',
                              na_values=['M'])
teaching_feeding.head()


# In[24]:


teaching_anthro = pd.read_csv(DATA_PATH + 'ncs_teaching_anthro_v1.csv',
                             na_values=['M'])
teaching_anthro.tail()


# In[25]:


teaching_child = pd.read_csv(DATA_PATH + 'ncs_teaching_child_v1_1.csv', 
                            index_col=0, na_values=['M'])
assert teaching_child.index.is_unique
teaching_child.tail()


# In[16]:


teaching_childhealth = pd.read_csv(DATA_PATH + 'ncs_teaching_childhealth_v1.csv',
                                  na_values=['M'])
teaching_childhealth.head()


# In[17]:


teaching_mompreghealth = pd.read_csv(DATA_PATH + 'ncs_teaching_mompreghealth_v1.csv',
                                    index_col=0, na_values=['M'])
assert teaching_mompreghealth.index.is_unique
teaching_mompreghealth.head()


# In[18]:


teaching_sleep = pd.read_csv(DATA_PATH + 'ncs_teaching_sleep_v1.csv',
                            na_values=['M'])
teaching_sleep.head()


# ## Data Visualizations

# In[20]:


pd.plotting.scatter_matrix(teaching_anthro, figsize=(14, 14));


# In[45]:


pd.plotting.scatter_matrix(teaching_sleep, figsize=(14, 14), alpha=0.1);


# In[41]:


plot_cols = ['CHILD_RACE','GESTATIONAL_AGE',
       'BABY_WEIGHT', 'MULTIPLE', 'SIBLINGS', 'WITHDREW', 
       'MOM_RACE', 'MOM_ETHNICITY', 'MOM_MARISTAT', 'MOM_EDUCATION',
       'MOM_INSURANCE', 'SURVEY_LANG', 'HOUSEHOLD_INCOME']


# In[42]:


ax = pd.plotting.parallel_coordinates(teaching_child[plot_cols].dropna(subset=['CHILD_RACE']), 
                                 'CHILD_RACE')
plt.xticks(rotation=90);