Array

The Endia Array is the main data structure for storing and manipulating multi-dimensional data.

fn __init__(inout self, shape: List[Int], requires_grad: Bool = False, is_complex: Bool = False)

Args

• shape (List[Int]): The shape of the array.
• requires_grad (Bool): Whether to track gradients for this array. (optional, default: False)
• is_complex (Bool): Whether the array contains complex numbers. (optional, default: False)

Endia Arrays behave in a similar way to PyTorch Tensors or Jax ndarrays.

Array creation

• array: Create an array from a string or a list
• zeros: Create an array filled with zeros with a given shape
• randu: Create an array filled with random numbers from a uniform distribution
• randn: Create an array filled with random numbers from a normal distribution
• full: Create an array filled with a scalar value and a given shape
• complex: Create a complex array from two real arrays

Methods

• Unary Operations: Trigonometric, exponential, logarithmic, ...
• Binary Operations: Addition, multiplication, division, ...
• Comparison Operations: Equal, greater, less, ...
• Reduce Operations: Sum, mean, ...
• View Operations: Reshape, slice, permute, ...
• Spacial Operations: Convolution, pooling, ...
• Stacking and Concatenation: Concatenate, pad, ...
• backward: Compute the gradients of the array with respect to the input arrays

See the complete list of operations here.

Example

import endia as nd
 
# Create a 2x2 array
a = nd.array('[[1, 2], [3, 4]]')
b = nd.array('[[5, 6], [7, 8]]')
 
# Compute the sum of the arrays
c = a + b
 
# Print the result
print(c)
 
# Output:
# Array([[6., 8.],
#        [10., 12.]], shape=(2, 2), dtype=float3232)