simpleflow

simpleflow copied to clipboard

SimpleFlow

A simple TensorFlow-like graph computation framework in Python for learning purpose

Quick Start

import simpleflow as sf

# Create a graph
with sf.Graph().as_default():
    a = sf.constant(1.0, name='a')
    b = sf.constant(2.0, name='b')
    result = sf.add(a, b, name='a+b')

    # Create a session to run the graph 
    with sf.Session() as sess:
        print(sess.run(result))

Examples

• Feedforward
• Linear Regression

Features

• [x] Computational Graph
• [x] Feed forward propagation
• [x] Backpropagation
• [x] GradientDescent Optimizer
• [x] Linear Regression Example
• [ ] MNIST classification Example

Blogs

• 实现属于自己的TensorFlow(一) - 计算图与前向传播
• 实现属于自己的TensorFlow(二) - 梯度计算与反向传播
• 实现属于自己的TensorFlow(三) - 反向传播与梯度下降实现