Blog


STM32 Project Ideas

posted Mar 2, 2021, 8:41 PM by MUHAMMAD MUN`IM AHMAD ZABIDI

Detect dogs and cats sounds in your neighborhood!
How screen scraping and TinyML can turn any dial into an API


MKEL1123 STM32 Projects

posted Feb 28, 2021, 6:19 AM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Feb 28, 2021, 6:25 AM ]


Group 8


Group 10


Kaggle is a Python Goldmine

posted Feb 26, 2021, 2:30 AM by MUHAMMAD MUN`IM AHMAD ZABIDI

Compeitions:
Python Notebooks on Spectrograms
Not on Kaggle, but related:
Haytham Fayek offered this advice:

tl;dr:
Use Mel-scaled filter banks if the machine learning algorithm is not susceptible to highly correlated input.
Use MFCCs if the machine learning algorithm is susceptible to correlated input.

Decoding MP3 files

posted Jan 23, 2021, 10:47 PM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Jan 23, 2021, 11:08 PM ]

  • [britanak2001efficient] Vladimir Britanak and K. R. Rao, An Efficient Implementation of the Forward and Inverse MDCT in MPEG Audio Coding, IEEE Sig Proc Letters, 2001.
  • [britanak2011survey] Vladimir Britanak, A survey of efficient MDCT implementations in MP3 audio coding standard: Retrospective and state-of-the-art, Signal Processing, 2011,
  • [raissi2002theory] Rassol Raissi, The Theory Behind Mp3, , December 2002.
  • [jacaba2001audio] Audio Compression Using Modified Discrete Cosine Transform: The Mp3 Coding Standard, 2001.
  • [da2008mdct] Xingdong Da, An MDCT Hardware Accelerator for MP3 Audio, 2008
  • [lee] Wen-Cheh Lee, Design of the Audio Coding Standards for MPEG and AC-3.
  • [kok1997fast] C.W. Kok, Fast Algorithm for Computing Discrete Cosine Transform, IEEE Trans Sig Proc, 1997: 
  • [cheng2003fast] Mu-Huo Cheng and Yu-Hsin Hsu, Fast IMDCT and MDCT Algorithms— A Matrix Approach, IEEE Trans Sig Proc, 2003:
  • [lee2001improved] Improved Algorithm for Efficient Computation of the Forward and Backward MDCT in MPEG Audio Coder, Szu-Wei Lee, IEEE Trans Cir Sys, 2001
  • Bibliography at MP3-Tech
  • Bibliography at CCRMA Stanford
  • [levine1998audio] Scott N. Levine, Audio Representations For Data Compression And Compressed Domain Processin, PhD Thesis, Stanford U, 1998.

STM32 in Academic Publications (Nov 2020)

posted Nov 27, 2020, 11:32 PM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Apr 13, 2021, 4:44 PM ]

Problem categories that benefit from machine learning

posted Nov 15, 2020, 2:09 AM by MUHAMMAD MUN`IM AHMAD ZABIDI

Fig. 2 from R. Boutaba et al., “A comprehensive survey on machine learning for networking: evolution, applications and research opportunities,” J. Internet Serv. Appl., vol. 9, no. 1, p. 99, 2018.

Problem categories that benefit from machine learning.
a Clustering.
b Classification.
c Regression.
d Rule extraction



The constituents of ML-based solutions

posted Nov 15, 2020, 2:06 AM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Nov 15, 2020, 2:06 AM ]

Fig. 3 from R. Boutaba et al., “A comprehensive survey on machine learning for networking: evolution, applications and research opportunities,” J. Internet Serv. Appl., vol. 9, no. 1, p. 99, 2018.

The evolution of machine learning techniques with key milestones

posted Nov 15, 2020, 1:59 AM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Nov 15, 2020, 2:03 AM ]

Fig. 5 from R. Boutaba et al., “A comprehensive survey on machine learning for networking: evolution, applications and research opportunities,” J. Internet Serv. Appl., vol. 9, no. 1, p. 99, 2018.


Computing devices that have been used for machine learning at the edge

posted Nov 15, 2020, 1:56 AM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Nov 15, 2020, 1:57 AM ]

Table 3 from M. G. S. Murshed, C. Murphy, D. Hou, N. Khan, G. Ananthanarayanan, and F. Hussain, “Machine Learning at the Network Edge: A Survey,” arXiv Prepr. arXiv1908.00080, pp. 1–28, 2019.

Device

GPU

CPU

RAM

Flash

memory

Power

consumption

Example

applications

Raspberry Pi

400MHz

VideoCore IV

Quad

Cortex A53 @ 1.2GHz

1 GB SDRAM

32 GB

2.5 Amp

video analysis [78, 114]

Coral Dev Board (Edge TPU)

GC7000 Lite

Graphics + Edge TPU coprocessor

Quad Cortex-A53, Cortex-M4F

1 GB LPDDR4

8 GB LPDDR4


5V DC

image processing [18]


SparkFun Edge


-

32-bit ARM

Cortex-M4F 48MHz

(with 96MHz

burst mode) processor


384KB


1MB


6uA/MHz


speech recognition [31]


Jetson TX1

Nvidia

Maxwell 256 CUDA

cores

Quad ARM A57/2 MB L2

4 GB

64 bit LPDDR4

25.6 GB/s

16 GB eMMC, SDIO, SATA


10-W

video, image analysis [68] [61]

robotics [33]


Jetson TX2

Nvidia Pascal

256 CUDA

cores

HMP Dual

Denver 2/2 MB L2 +

Quad ARM

A57/2 MB L2

8 GB

128 bit LPDDR4

59.7 GB/s


32 GB eMMC, SDIO, SATA


7.5-W


video, image analysis [68], [92]

robotics [24]

Intel Movidius Neural Compute Stick


High Performance VPU

Myriad 2 Vision Processing Unit


1 GB


4 GB

2 trillion

16-bit operations per second within

500 mW


classification [73] computer

vision [14, 43]

ARM ML

-

ARM ML

processor

1 GB

-

4 TOPs/W

(Tera Operations)

image, voice recognition [107]


RISC-V GAP8


-

nona-core

32-bit RISC-V

microprocessor

@250 MHz


16 MiB SDRAM


-


1 GOPs/mW


image, audio processing [35]

OpenMV

Cam

-

ARM 32-bit

Cortex-M7

512KB

2 MB

200mA

@ 3.3V

image

processing [6]

BeagleBone AI

-

Cortex-A15

Sitara AM5729 SoC with 4 EVEs

1 GB

16 GB

-

computer vision [25]

EMC3531

-

ARM Cortex-M3

NXP Coolflux DSP

-

-

-

audio, video analysis38

Machine learning frameworks that have been used on edge devices

posted Nov 15, 2020, 1:53 AM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Nov 15, 2020, 1:55 AM ]

Table 3 from M. G. S. Murshed, C. Murphy, D. Hou, N. Khan, G. Ananthanarayanan, and F. Hussain, “Machine Learning at the Network Edge: A Survey,” arXiv Prepr. arXiv1908.00080, pp. 1–28, 2019.

Framework

Core language

Interface

Part running on the edge

Example applications

TensorFlow Lite (Google)

C++

Java

C/C++

Java

TensorFlow Lite NN API

computer vision [109],

speech recognition [42, 1]

Caffe2

Caffe2Go (Facebook)

C++

Android iOs

NNPack

image analysis, video analysis [53]

Apache MXNet

C++

Python R

Linux

MacOS Windows

Full Model

object detection, recognition [78]

Core ML2 (Apple)

Python

iOS

CoreML

image analysis [16]

NLP [105]

ML Kit (Google)

C++

Java

Android iOs

Full Model

image recognition,

text recognition, bar-code scaning [26]

AI2GO

C, Python Java, Swift

Linux macOs

Full Model

object detection, classification [5]

DeepThings

C/C++

Linux

Full Model

object detection [119]

DeepIoT

Python

Ubilinux

Full Model

human activity

recognition,

user identification [116]

DeepCham

C++

Java

Linux Android

Full Model

object recognition [62]

SparseSep

-

Linux Android

Full Model

mobile object

recognition, audio classification [15]

Edgent

-

Ubuntu

Major part of the DNN

image recognition [63]


17

1-10 of 166

Comments