Skip to content
Snippets Groups Projects
Commit 31db043e authored by Hammouda Elbez's avatar Hammouda Elbez :computer:
Browse files

SpiNNaker: ReadMe updated

parent e65835e8
No related branches found
No related tags found
No related merge requests found
## Files:
You can change and use the file names as you prefer. As an example, here we did it as below:
Network: Input(784)*FC1(400)*FC2(1600)
weights_fc1_X.Y:
Weight-Files from Input_layer to FC1 layer
weights_fc2_X.Y:
Weight-Files from FC1 to FC2 layer
X: baseline, compressed
Y:
nothing: binary file from SCNN
txt: generated connection file for spynnaker (PyNN for SpiNNaker)
## Files naming
In order to properly use the provided scripts, make sure to rename the binary weights extracted from the csnn-simulator as follows:
As an example, we have the next architecture with MNIST dataset:
`Input(784) x Fc1(400) x Fc2(1600)`
weights_fc1_X:
Weights between Input and Fc1 layer
weights_fc2_X:
Weights between Fc1 and Fc2 layer
X = type of network [baseline or compressed]
>`binary file` : are from the csnn-simulator (Weight Extractor), which <span style="color:red;">needs to be renamed by the user</span>
>`.txt file` : represents the generated connections file for Spynnaker
\ No newline at end of file
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment