Skip to content
Snippets Groups Projects
Select Git revision
  • full_accl_support
  • georg/accl-build
  • georg/accl-finn
  • georg/end2end
  • adegendt/accl_clone
  • accl default
  • georg/accl-complete
  • main protected
  • dev
9 results
Blame Permalink
demos.md 4.50 KiB

Neural Network Demos

Multiple Jupyter notebooks examples are provided, with different datasets and two architectures:

  • Feed-forward Dataflow: all layers of the network are implemented in the hardware, the output of one layer is the input of the following one that starts processing as soon as data is available. The network parameters for all layers are cached in the on-chip memory. For each network topology, a customized hardware implementation is generated that provides low latency and high throughput.

  • Dataflow with loopback: a fixed hardware architecture is implemented, being able to compute multiple layers in a single call. The complete network is executed in multiple calls, which are scheduled on the same hardware architecture. Changing the network topology implies changing the runtime scheduling, but not the hardware architecture. This provides a flexible implementation but features slightly higher latency.

Our design examples are mostly for the PYNQ Z1 and Z2 boards, and a few for the Ultra96. Future support for AWS F1 and other Xilinx platforms is also planned.

Demos with Dataflow Architecture

Thumbnail Dataset Neural Network Task Link
drawing CIFAR-10 6 convolutional, 3 max pool and 3 fully connected layers Image classification (animals and vehicles) Cifar10
drawing Street View House Numbers 6 convolutional, 3 max pool and 3 fully connected layers Image classification (house numbers) SVHN
drawing German Road Signs 6 convolutional, 3 max pool and 3 fully connected layers Image classification (road signs) GTRSB
drawing MNIST 3 fully connected layers Image classification (handwritten digits) MNIST
drawing Fraktur Bi-LSTM Optical Character Recognition Fraktur

Demos with Loopback Architecture

  • ImageNet Classification: shows an example on how to classify a non-labelled image (e.g., downloaded from the web, your phone etc) in one of the 1000 classes available on the ImageNet dataset.

  • ImageNet - Dataset validation: shows an example classifying labelled image (i.e., extracted from the dataset) in one of the 1000 classes available on the ImageNet dataset.

  • ImageNet - Dataset validation in a loop: shows an example classifying labelled image (i.e., extracted from the dataset) in one of the 1000 classes available on the ImageNet dataset in a loop.

  • Object Detection - from image: shows object detection in a image (e.g., downloaded from the web, your phone etc), being able to identify objects in a scene and drawing bounding boxes around them. The objects can be one of the 20 available in the PASCAL VOC dataset