Show page Old revisions

Differences

This shows you the differences between two versions of the page.

--- graphgencnn [2014/11/04 20:31]
gweisz
+++ graphgencnn [2015/01/20 14:40] (current)
gweisz
@@ Line 1: / Line 1: @@
-====== GraphGen-based Convolutional Neural Network for the ZedBoard ======
+==== GraphGen-based Convolutional Neural Network for the ZedBoard ====
 This page hosts a Convolutional Neural Network demo for the [[http://http://zedboard.org/|ZedBoard]] built using the [[http://www.ece.cmu.edu/~coram/lib/exe/fetch.php?media=http%3A%2F%2Fwww.ece.cmu.edu%2F~coram%2Fpublications%2Fgraphgen.pdf|GraphGen compiler]].
@@ Line 16: / Line 16: @@
 ===== The Demo =====
-This neural network demo is a handwriting recognition application that uses a [[http://www.codeproject.com/Articles/16650/|publicly available]] neural network that recognizes handwritten digits from the [[http://www.codeproject.com/Articles/16650/Neural-Network-for-Recognition-of-Handwritten-Digi#AboutMNist|MNIST data set]]. In our 2014 FCCM paper referenced above, we show results of accelerating this neural network on the Xilinx ML605 FPGA and Terasic DE4 FPGA using an Altera chip. This demo demonstrates the neural network on the [[http://www.zedboard.org|ZedBoard]], an inexpensive board that includes the Xilinx Zynq 7020 FPGA, a System-on-Chip that also includes two ARM cores. The demo uses software running on an ARM core to support host communications through the ZedBoard's ethernet interface, and communicate with the GraphGen-created processing system through an AXI slave interface. The processing system connects to the ZedBoard's DRAM through the Zynq chip's high performance AXI ports. This demo works on the original ZedBoard, and does not support the MicroZed, which uses a smaller FPGA.
+This neural network demo is a handwriting recognition application that uses a [[http://www.codeproject.com/Articles/16650/|publicly available]] neural network that recognizes handwritten digits from the [[http://www.codeproject.com/Articles/16650/Neural-Network-for-Recognition-of-Handwritten-Digi#AboutMNist|MNIST data set]]. In our 2014 FCCM paper, referenced above, we show results of accelerating this neural network on the Xilinx ML605 FPGA and Terasic DE4 FPGA using an Altera chip. This demo demonstrates the neural network on the [[http://www.zedboard.org|ZedBoard]], an inexpensive board that includes the Xilinx Zynq 7020 FPGA, a System-on-Chip that also includes two ARM cores. This is the same demo that we presented at Demo Night at FCCM 2014.
-This demo runs through the 10,000 test images that come with the neural network on the [[http://www.codeproject.com/Articles/16650/|CodeProject]] site.
+The demo uses software running on an ARM core to support host communications through the ZedBoard's ethernet interface, and communicate with the GraphGen-created processing system through an AXI slave interface. The processing system connects to the ZedBoard's DRAM through the Zynq chip's high performance AXI ports. This demo works on the original ZedBoard, and does not support the MicroZed, which uses a smaller FPGA.
+This demo runs through the 10,000 test images that come with the neural network on the [[http://www.codeproject.com/Articles/16650/|CodeProject]] site. The demo runs through the test images, and allows the user to control the run time of the application by selecting a "small" subgraph size, where a single output neuron is calculated for each subgraph, or a "large" subgraph size, which uses the largest subgraphs that can fit on the ZedBoard. Larger subgraphs allow for more efficient computation because of data reuse among points in the subgraph.
 === Running the demo ===
@@ Line 26: / Line 29: @@
 == Demo Instructions ==
-  - Configure the ZedBoard. Copy "boot.bin" from the archive onto an SD card, and put the card into the ZedBoard. Configure the ZedBoard to boot off of the SD card by setting jumpers M05 and M04 to 1, and M03 to 0, as shown in the picture below.
+. Configure the ZedBoard. Copy "boot.bin" from the archive onto an SD card, and put the card into the ZedBoard. Configure the ZedBoard to boot off of the SD card by setting jumpers M05 and M04 to 1, and M03 to 0, as shown in the picture below.
 {{:zed-jumpers-sd.png?200|}}
-  - Turn on the ZedBoard, and connect it to the host computer through an Ethernet cable. The ZedBoard will be configured to IP 192.168.1.10, so the host computer must be configured to be able to communicate with this IP
-  - Wait for the ZedBoard to finish booting. The ZedBoard has booted when the blue "done" LED to the left of the OLED display (near the bottom corner) comes on.
+. Turn on the ZedBoard, and connect it to the host computer through an Ethernet cable. The ZedBoard will be configured to IP 192.168.1.10, so the host computer must be configured to be able to communicate with this IP.
-  - Start the host control program. The host program will show the following screen:
+. Wait for the ZedBoard to finish booting. The ZedBoard has booted when the blue "done" LED to the left of the OLED display (near the bottom corner) comes on.
+. Start the host control program. The host program will show the following screen:
 {{:demo-shot1.png|}}
+The UI for the host control program has four sections:
+    * The top left section has a control panel with several controls:
+      * A "Next" button allowing the user to advance to the next image in the data set
+      * A "Run" button that will cycle through all the images in the data set as quickly as possible
+      * "Small" and "Large" buttons, allowing the user to select "Small" or "Large" subgraph sizes for computation. The small subgraph size calculates a new value for a single neuron per subgraph, and the larger subgraphs calculate on the largest subgraphs that fit on the ZedBoard
+      * A "Mean run times" panel. showing the average run times for small and large subgraph sizes
+    * The top left image shows the current image to recognize.
+    * The bottom right image shows the recognition result for the current image, if available. It will be blank before the current image has been recognized
+    * The bottom left image shows the label for the current image. The first image that the application loads is not part of the labeled data set, so will not display a label.
+. Press the button labeled "Large." This will cause the application to run, and show a recognition result in the bottom right pane. It will also show the run time of the application.
+. Press the button labeled "Small." This will run the application with the small subgraph size, which takes significantly longer than the large subgraph size.
+. Press the button labeled "Large", and press the "Run" button. The application will now cycle through the sample images. Note that as the application recognizes labeled images, the bottom right pane will show a correctly recognized result with a green background, and an incorrectly recognized result with a red background. Not also that the label for the current image is now shown on the bottom left pane, and the overall recognition accuracy rate for the neural network is updated in real time as the data is recognized. This neural network does get over 95% accuracy on the entire data set, but it is expected that there will be some errors.
+Please contact [[http://www.google.com/recaptcha/mailhide/d?k=0111talvydAJhFRTlfF0Bknw==&c=RvEGAU-BIM-FCSh7I3xUm-oxXANpgfPiBWqxT8IcLYA=|Gabe Weisz]] with any questions.
+==Acknowledgements==
+Funding for this work has been provided, in part, by the National Science Foundation (CCF-1320725) and the Intel Science and Technology Center on Embedded Computing.

Back to top

graphgencnn.1415133118.txt.gz · Last modified: 2014/11/04 20:31 by gweisz

Sitemap Recent Changes