Deep Learning Introduction: Samples

These examples have been developed for the course "Deep Learning Introduction", taught by Francisco M. Castro and Paula Ruiz Barroso.

How to use

Clone this repository with git.

git clone https://github.com/fmcp/jetson_uma

Maybe neccesary:

sudo pip3 install --global-option=build_ext --global-option="-I/usr/local/cuda/include" --global-option="-L/usr/local/cuda/lib64" pycuda

Import each *.ipynb file contained in Notebooks folder into a different Google Colab Notebook.

Proposed projects

• Identification of bottles with different drinks
• Identification of playing card suits
• Recognition of handwritten music notes
• Counting people in a room
• Facial emotion recognition

Network routing for Linux Hosts

1. Connect your Jetson Nano to your computer using the Micro-USB cable.
2. Open your Network Manager and check if the wired connection is restricted to enp60s0
3. Enable IP forwarding echo 1 > /proc/sys/net/ipv4/ip_forward
4. Add the following routing rule for iptables: sudo iptables -t nat -A POSTROUTING -o XXX -j SNAT --to YYY where XXX is the interface used for your internet connection and YYY is the IP assigned to your computer. For example, for Wifi connections: iptables -t nat -A POSTROUTING -o wlo0 -j SNAT --to 172.18.14.155
5. Connect to the Jetson Nano: ssh [email protected]

Authors

• Francisco M. Castro - fmcp ([email protected])
• Paula Ruíz Barroso - PaulaRuizB ([email protected])

License

This project is licensed under the MIT License - see the LICENSE.md file for details