@@ -7,20 +7,27 @@ VTA(versatile tensor accelerator) is an open-source deep learning accelerator st
77It is not just an open-source hardware, but is an end to end solution that includes
88the entire software stack on top of VTA open-source hardware.
99
10-
1110The key features include:
1211
1312- Generic, modular open-source hardware
1413 - Streamlined workflow to deploy to FPGAs.
15- - Simulator support
16- - Driver and JIT runtime for both simulated backend and FPGA.
14+ - Simulator support to protoype compilation passes on regular workstations.
15+ - Driver and JIT runtime for both simulated and FPGA hardware backend .
1716- End to end TVM stack integration
1817 - Direct optimization and deploy models from deep learning frameworks via TVM stack.
19- - Customized and extendible TVM compiler backend
20- - Flexible RPC support to ease the deployment, you can program it with python :)
18+ - Customized and extendible TVM compiler backend.
19+ - Flexible RPC support to ease the deployment, and program FPGAs with Python
2120
2221VTA is part of our effort on [ TVM Stack] ( http://www.tvmlang.org/ ) .
2322
23+ VTA Installation
24+ ----------------
25+ To get started with VTA, please follow the [ Installation Guide] ( docs/how_to/install.md )
26+
27+ ResNet-18 Inference Example
28+ ---------------------------
29+ To offload ResNet-18 inference, follow the [ ResNet-18 Guide] ( examples/resnet18/pynq/README.md )
30+
2431License
2532-------
2633© Contributors, 2018. Licensed under an [ Apache-2.0] ( https://github.com/tmoreau89/vta/blob/master/LICENSE ) license.
0 commit comments