CFU Playground (1) - Introduction

-

It is not an official open source project of Google, but a very interesting project is in progress. (Refer to the site below for details)

https://cfu-playground.readthedocs.io/en/latest/index.html

To explain the concep briefly, it is a concept that puts a RISC-V soft-core in the FPGA and allows ML to run on the soft-core. The interesting point is that we can easily customize the core with accelerators on FPGA and also add more custom instructions. CFU playgound framework is composed of many good opensource products. So, it's free. We can practice it through many projects of Pete Warden, who devised TinyML using Tensorflow lite micro Framework a few years ago. And we can refer a few more interesting projects. 

One of important enablers for CFU playground would be Tensorflow lite micro in conjuction with TinyML. The purpose of TinyML is to implement ML using only a few hundred KByte of memory in an ARM Cortex-M-class microcontroller. Tensorflow lite is already used a lot by people who develop mobile applications, and Tensorflow lite micro can be completed by applying 8-bit quantization there to make machine learning algorithms run on it. (see link below)

https://www.tensorflow.org/lite/microcontrollers?hl=en

The purpose of CFU Playground is to implement an accelerator that improves the performance of machine learning by adding logics to microcontrollers. In the conventional CPU, ALU (Arithmetic Logic Unit) performs operations together with designated registers, where CFU is added to enable faster execution of operations suitable for ML. By following the tutorial on the site above, installing various tools on Linux, and targeting the recommended board, you can better understand the overall flow. Here, even if there is no board, there is an open source program called renode that emulates the board, so the ARTY-7 FPGA board can be completely driven. You don't need to purchase a separate board for the tutorial. If you want to implement a more practical example in the future, you may need to obtain a board suitable for the purpose. It's a really fun concept, so I recommend you try it. 

In the previous RNN project, I try to minimize CPU intervention by implementing a full custom ML accelerator. In the future project, I will implement the same ML algorithm in the CFU Playground, understand how the entire tool chain works, and examine the strengths and weaknesses of the previous project. 

1. Implementation of RNN algorithm with Tensorflow lite micro: Since CFU Playground's basic python language is Tensorflow. So, I need to convert previously implemented Pytorch model into Tensorflow model. 

2. After securing the ML C source using Tensorflow lite micro, check if it operates in the RISC-V environment and figure out which part can be accelerated to improve performance. 

3. Based on the above analysis, implement the accelerator and check if there is any performance improvement. I plan to post the above three steps on the blog as they progress.

Comments

Post a Comment

Popular posts from this blog

RNN (5) - FPGA System Design

-
Image
This is the last post on hardware implementation. The RTL codes of the previous post were designed for HW Accelerator IP to perform RNN, and this post describes how to design the whole system for targetting FPGA. The FPGA board used here is an FPGA Board using Xilinx ZYNQ-7000 as open hardware called Pynq-z2. (For Pynq-z2, please refer to the link below.) -  http://www.pynq.io/ -  https://www.tulembedded.com/FPGA/ProductsPYNQ-Z2.html -  https://pynq.readthedocs.io/en/latest/ First, the RNN Accelerator must open a channel to communicate with the ARM-Cortex A9 where SW is running. IP Packaging is performed using Vivado of Xilinx, and I used the AXI interface template provided by Vivado. RTL codes  on the right of the figure below  in the code window  shows how to  combine RNN IP with AXI interface. IP Packaging is possible easily through a few steps in Vivado. (Details will be posted separately.) After IP Packaging, you need to create a bit file to downl...
Read more

RNN (3) - System Simulator

-
Image
The Pytorch reference in the previous post uses the optimized library and floating point number, so the performance can be considered the most ideal. In the system simulator for Verilog design, some restrictions follow.  - Fixed Point Number: Because the range of numbers is limited, performance restrictions follow.  - Suboptimal Matrix Operation: The existing Pytorch model has been optimized for efficient and accurate operation in the Python environment. There is a difference in performance in the System Simulator that is actually implemented.  - Non-linear activation function: Here, tanh (hyperbolic tangent) is used, and linear approximation is performed for convenience of implementation.  Here, a 16-bit fixed point number is used. In fact, all 16 bits are not used and only 7 decimal places are used for easy calculation. Unused redundant bits should be optimized and removed in the future, but the truncation so that prevent overflow during every Addition/Multiplicati...
Read more

RNN (4) - RTL Design

-
Image
I designed the RTL in the same way as the System Simulator designed in the previous post. The major considerations for RTL design are as follows.  (The detailed code description will be posted in a separate post later.  Also, there are many good tutorials on how to use Intel Quartus Modelsim, so you'd better search for it. I will post the explanation of simulating this code in a separate post. ) - I/O method: The most common method is to save the input in the memory, drive the RNN, save the predicted value to output memory, and then send the completion signal. At this time, the memory can be selected in various ways. Various methods such as FIFO, Single Port, Dual Port, etc. can be used. Here, Single Port Memory is selected as the simplest.  - Timing: RTL stands for Register Transfer Level. I don't know who put it on, but it seems to have been put together relatively well. Register is a component whose output is changed at the edge of the clock. Since there is some logic ...
Read more