| Title: | Towards deploying highly quantized neural networks on FPGA using chisel |
|---|
| Authors: | ID Vreča, Jure, Institut Jožef Stefan (Author)
ID Biasizzo, Anton, Institut Jožef Stefan (Author) |
|---|
| Files: |  URL - Source URL, visit https://ieeexplore.ieee.org/document/10456782/authors#authors
 PDF - Presentation file, download (419,83 KB)
MD5: C6531548B85A232A34C012D82828C084
|
|---|
| Language: | English |
|---|
| Typology: | 1.08 - Published Scientific Conference Contribution |
|---|
| Organization: |  IJS - Jožef Stefan Institute
|
|---|
| Abstract: | We present chisel4ml, a Chisel-based tool that generates hardware for highly quantized neural networks described in QKeras. Such networks typically use parameters with bitwidths less than 8 bits and may have pruned connections. Chisel4ml can generate the highly quantized neural network as a single combinational circuit with pipeline registers in between the different layers. It supports heterogeneous quantization where each layer can have a different precision. The full parallelization enables very low-latency and high throughput inference, that are required for certain tasks. We illustrate this on the triggering system for the CERN Large Hadron Collider, which filters out events of interest and sends them on for further processing. We compare our tool against hls4ml, a high-level synthesis based approach for deploying similar neural networks. Chisel4ml is still under development. However, it already achieves comparable results to hls4ml for some neural network architectures. Chisel4ml is available on https://github.com/cs-jsi/chisel4ml. |
|---|
| Keywords: | neural networks, QKeras, Chisel4ml |
|---|
| Publication status: | Published |
|---|
| Publication version: | Author Accepted Manuscript |
|---|
| Publication date: | 19.03.2023 |
|---|
| Publisher: | IEEE |
|---|
| Year of publishing: | 2023 |
|---|
| Number of pages: | Str. 161-167 |
|---|
| Source: | ZDA |
|---|
| PID: | 20.500.12556/DiRROS-18802  |
|---|
| UDC: | 004 |
|---|
| DOI: | 10.1109/DSD60849.2023.00032 |
|---|
| COBISS.SI-ID: | 190218499 |
|---|
| Copyright: | ©2023 IEEE |
|---|
| Note: | Nasl. z nasl. zaslona;
Opis vira z dne 25. 3. 2024;
|
|---|
| Publication date in DiRROS: | 23.04.2024 |
|---|
| Views: | 383 |
|---|
| Downloads: | 244 |
|---|
| Metadata: |    |
|---|
|
:
|
Copy citation |
|---|
| | | | Share: |  |
|---|
Hover the mouse pointer over a document title to show the abstract or click
on the title to get all document metadata. |
