Skip to content

Accelerator for a Convolutional Layer modeled in matlab, and implemented <64,7> in verilog

License

Notifications You must be signed in to change notification settings

imthil/ConvLayerAccelerator

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

33 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Accelerating the Convolutional Layer

Accelerator for a Convolutional Layer written in Verilog. For SCU ELEN249, taught by Kiran Gunnam.

Deliverables

  1. Implement all the 3 approaches listed in the paper (Code 1, Code 2 and Code 3) in MATLAB.

    completed in ./matlab/ folder

  2. Model the core part of the hardware accelerator (as shown in Fig 7 of the paper for the convolutional layer with unroll factor <64,7> in MATLAB and Verilog . Both should match.

    completed in ./64_7/ folder

  3. Model the DRAM access patterns for the input and output for all the 3 approaches in MATLAB.

    see README.md in ./matlab/ folder

About

Accelerator for a Convolutional Layer modeled in matlab, and implemented <64,7> in verilog

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published