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

# # Digit Recognizer using Random Forest

# * https://www.kaggle.com/c/digit-recognizer

# In[1]:


import pandas as pd
import numpy as np
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split


# ## Prepare Data

# In[3]:


train = pd.read_csv('../input/digit-recognizer/train.csv')
test  = pd.read_csv('../input/digit-recognizer/test.csv')
train.shape, test.shape


# In[6]:


images = train.iloc[:,1:]
labels = train.iloc[:,:1]
images.shape, labels.shape, test.shape


# In[14]:


train_images, valid_images, train_labels, valid_labels = train_test_split(images, labels, train_size=0.8, test_size=0.2, random_state=0)
train_images.shape, valid_images.shape


# ## Build Model

# In[15]:


clf = RandomForestClassifier(random_state=0)
clf.fit(train_images.values, train_labels.values.ravel())


# In[18]:


clf.score(valid_images, valid_labels)


# ## Submit

# In[19]:


predictions = clf.predict(test)


# In[20]:


submissions = pd.DataFrame({
    "ImageId": list(range(1, len(predictions)+1)),
    "Label": predictions})
submissions.to_csv("output.csv", index=False, header=True)


# In[ ]: