Diabetic Prediction

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

# # DIABETIC PREDICTION

# ## IMPORTING THE REQUIRED DEPENDENCIES

# In[1]:


import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from sklearn import svm
from sklearn.metrics import accuracy_score


# ### EXPLORING AND ANALYSING THE DATASET

# In[2]:


dp=pd.read_csv('D:\Datasheet\diabetes.csv')


# In[3]:


dp.head()


# In[4]:


dp.tail()


# In[5]:


dp.isnull().sum()


# In[6]:


dp.describe()


# In[7]:


dp.groupby('Outcome').mean()


# In[8]:


dp['Outcome'].value_counts()


# ### OBSERVING THE DATA USING VISUALS

# In[9]:


sns.countplot(x='Outcome',data=dp)
sns.set(font_scale=0.1)
sns.set(rc={'figure.figsize':(30,20)})
plt.show()


# ### SPLITTING OF DATA

# In[60]:


X=dp.drop(columns='Outcome',axis=1)
Y=dp['Outcome']


# In[61]:


X


# In[62]:


Y


# ### DATA STANDARDIZATION

# In[63]:


sc=StandardScaler()


# In[64]:


sd=sc.fit_transform(X)


# In[65]:


sd


# In[66]:


X=sd


# ### TRAIN TEST DATA SPLIT

# In[67]:


X_tr,X_t,Y_tr,Y_t=train_test_split(X,Y,test_size=0.2,stratify=Y, random_state=2)


# In[68]:


print(X_tr)


# In[69]:


print(X_t)


# In[70]:


print(Y_tr)


# In[71]:


print(Y_t)


# ### TRAINING THE MODEL

# In[72]:


cf=svm.SVC(kernel='linear')


# In[73]:


cf.fit(X_tr,Y_tr)


# ### TESTING THE MODEL 
# #### OVER TRAINING DATA

# In[74]:


X_tr_p=cf.predict(X_tr)


# In[75]:


tr_accuracy=accuracy_score(X_tr_p,Y_tr)


# #### ACCURACY SCORE

# In[76]:


tr_accuracy


# #### OVER TEST DATA

# In[77]:


X_t_p=cf.predict(X_t)


# In[78]:


t_accuracy=accuracy_score(p,Y_t)


# #### ACCURACY SCORE

# In[79]:


t_accuracy


# ### PREDICTIVE SYSTEM
# 

# In[82]:


input_l=(6,148,72,35,0,33.6,0.627,50)


# In[85]:


input_a= np.asarray(input_l)
input_r=input_a.reshape(1,-1)
std=sc.fit_transform(input_r)
print(std)


# In[88]:


pre=cf.predict(std)


# In[89]:


if(pre==0):
    print('Person is non diabetic')
elif(pre==1) : 
    print("Person is diabetic")
else:
    print("Wrong Input")


# In[ ]: