This is a short post that explains how to write a high-performance matrix multiplication program on modern processors. In this tutorial I will use a single core of the Skylake-client CPU with AVX2, but the principles in this post also apply to other processors with different instruction sets (such as AVX512).
Matrix multiplication is a mathematical operation that defines the product of
##VGG16 model for Keras
This is the Keras model of the 16-layer network used by the VGG team in the ILSVRC-2014 competition.
It has been obtained by directly converting the Caffe model provived by the authors.
Details about the network architecture can be found in the following arXiv paper:
Very Deep Convolutional Networks for Large-Scale Image Recognition
K. Simonyan, A. Zisserman
| #!/usr/bin/env python | |
| """ | |
| Example of using Keras to implement a 1D convolutional neural network (CNN) for timeseries prediction. | |
| """ | |
| from __future__ import print_function, division | |
| import numpy as np | |
| from keras.layers import Convolution1D, Dense, MaxPooling1D, Flatten | |
| from keras.models import Sequential |
| import re | |
| _comparer = None | |
| import contextlib | |
| #ALWAYS_INLINE = "__attribute__((__always_inline__))" | |
| ALWAYS_INLINE = "ALWAYS_INLINE" | |
| class CodeGenerator(object): | |
| def __init__(self): |
| #/usr/bin/env python | |
| import math | |
| class RunningCorrelation: | |
| def __init__(self): | |
| self.n = 0.0 | |
| self.mean_x = 0.0 | |
| self.mean_y = 0.0 | |
| self.m2_x = 0.0 | |
| self.m2_y = 0.0 |
| /* | |
| A Minimal Capture Program | |
| This program opens an audio interface for capture, configures it for | |
| stereo, 16 bit, 44.1kHz, interleaved conventional read/write | |
| access. Then its reads a chunk of random data from it, and exits. It | |
| isn't meant to be a real program. | |
| From on Paul David's tutorial : http://equalarea.com/paul/alsa-audio.html |