ATtiny85-MNIST-RNN-EEPROM

ATtiny85 arduino example, running an RNN MNIST model via the (internal) 512-Byte EEPROM

Steps

• First ensure you have installed ATTinyCore & NeuralNetwork.h
• Second, save the RNN into the EEPROM via example
• Finally flash this

^{(PS. You will also need an indicator/led if you don't use a board such as DigiSpark)}

Testing

Do you want to test whether the results are real or not? run it natively on you OS directly, like:

git clone https://github.com/GiorgosXou/ATTiny85-MNIST-RNN-EEPROM
cd ./ATTiny85-MNIST-RNN-EEPROM/native_os_testing_example/
wget https://raw.githubusercontent.com/GiorgosXou/NeuralNetworks/refs/heads/master/src/NeuralNetwork.h
wget https://python-course.eu/data/mnist/mnist_train.csv
wget https://python-course.eu/data/mnist/mnist_test.csv
g++ -std=c++11 main.cpp -o main.o -O3 -march=native -ffast-math
./main.o

---

(See also some funny visualizations)