MartaAndronic / PolyLUT
PolyLUT is the first quantized neural network training methodology that maps a neuron to a LUT while using multivariate polynomial function learning to exploit the flexibility of the FPGA soft logic.
☆44Updated 9 months ago
Related projects ⓘ
Alternatives and complementary repositories for PolyLUT
- A collection of tutorials for the fpgaConvNet framework.☆32Updated 2 months ago
- Verilog implementation of Softmax function☆48Updated 2 years ago
- NeuraLUT: Hiding Neural Network Density in Boolean Synthesizable Functions☆13Updated 7 months ago
- SSR: Spatial Sequential Hybrid Architecture for Latency Throughput Tradeoff in Transformer Acceleration (Full Paper Accepted in FPGA'24)☆26Updated 4 months ago
- Open-source of MSD framework☆14Updated last year
- SAMO: Streaming Architecture Mapping Optimisation☆32Updated last year
- A systolic array simulator for multi-cycle MACs and varying-byte words, with the paper accepted to HPCA 2022.☆65Updated 3 years ago
- The codes and artifacts associated with our MICRO'22 paper titled: "Adaptable Butterfly Accelerator for Attention-based NNs via Hardware …☆113Updated last year
- Multi-core HW accelerator mapping optimization framework for layer-fused ML workloads.☆40Updated this week
- CHARM: Composing Heterogeneous Accelerators on Versal ACAP Architecture☆124Updated 2 weeks ago
- An HLS based winograd systolic CNN accelerator☆48Updated 3 years ago
- ☆70Updated last year
- Quantized ResNet50 Dataflow Acceleration on Alveo, with PYNQ☆52Updated 3 years ago
- ☆24Updated 8 months ago
- Implementation of Microscaling data formats in SystemVerilog.☆12Updated 2 months ago
- ☆24Updated 2 weeks ago
- [TCAD'23] AccelTran: A Sparsity-Aware Accelerator for Transformers☆33Updated last year
- FracBNN: Accurate and FPGA-Efficient Binary Neural Networks with Fractional Activations☆87Updated 3 years ago
- HLS implemented systolic array structure☆41Updated 7 years ago
- Performance and resource models for fpgaConvNet: a Streaming-Architecture-based CNN Accelerator.☆27Updated 2 weeks ago
- ☆60Updated 5 years ago
- ☆30Updated 2 months ago
- An Open Workflow to Build Custom SoCs and run Deep Models at the Edge☆65Updated this week
- FlexASR: A Reconfigurable Hardware Accelerator for Attention-based Seq-to-Seq Networks☆42Updated 2 years ago
- ☆32Updated 5 years ago
- ☆69Updated 4 years ago
- ☆55Updated 4 years ago
- An FPGA Accelerator for Transformer Inference☆73Updated 2 years ago
- A Reconfigurable Accelerator with Data Reordering Support for Low-Cost On-Chip Dataflow Switching☆30Updated last month
- A fast, accurate trace-based simulator for High-Level Synthesis.☆38Updated 6 months ago