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

# # Data Manipulation with Ndarray
# 
# Importing `np` (numpy-like) module and `npx` (numpy extensions) module from MXNet. 

# In[1]:


from mxnet import np, npx
# Invoke the experimental numpy-compatible feature in MXNet 
npx.set_np()  


# Create a vector and query its attributes

# In[2]:


x = np.arange(12)
x


# In[3]:


x.shape


# In[4]:


x.size


# More ways to construct arrays

# In[5]:


np.zeros((3, 4))


# In[6]:


np.array([[2, 1, 4, 3], [1, 2, 3, 4], [4, 3, 2, 1]])


# In[7]:


np.random.normal(0, 1, size=(3, 4))


# Elemental-wise operators

# In[8]:


x = np.array([1, 2, 4, 8])
y = np.ones_like(x) * 2
print('x =', x)
print('x + y', x + y)
print('x - y', x - y)
print('x * y', x * y)
print('x ** y', x ** y)
print('x / y', x / y)


# Matrix multiplication.

# In[9]:


x = np.arange(12).reshape((3,4))
y = np.array([[2, 1, 4, 3], [1, 2, 3, 4], [4, 3, 2, 1]])
np.dot(x, y.T)


# Concatenate arrays along a particular axis. 

# In[10]:


np.concatenate([x, y], axis=0), np.concatenate([x, y], axis=1)


# Broadcast Mechanism

# In[11]:


a = np.arange(3).reshape((3, 1))
b = np.arange(2).reshape((1, 2))
print('a:\n', a)
print('b:\n', b)
a + b


# Indexing and Slicing
# 

# In[12]:


print('x[-1] =\n', x[-1])
print('x[1:3] =\n', x[1:3])
print('x[1:3, 2:4] =\n', x[1:3, 2:4])
print('x[1,2] =', x[1,2])


# `mxnet.numpy.ndarray` and `numpy.ndarray`

# In[13]:


a = x.asnumpy()
print(type(a))
b = np.array(a)
print(type(b))