Neural Network from Scratch without ML libraries | 100 lines of Python code

preview_player
Показать описание
Implementation of a Neural Network from Scratch in 100 lines of Python code, without using AD libraries (PyTorch, TensorFlow, JAX, ...).

-----------------------------------------------------------------------------------------------------
#python #neuralnetworks #machinelearning #artificialintelligence #deeplearning #data #supervisedlearning #research #neural #function #relu #fromscratch #perceptron #multilayerperceptron #mlp #classification #MNIST #logsoftmax #nllloss
Рекомендации по теме
Комментарии
Автор

Quite à subir l'accent français, je préfèrerais l'entendre dans ma langue natale. 😄
I barely stand it in english, sorry.
Congratulations on the effort to synthesize so much into one short implementation and video. It quite demystifies machine learning.

emjizone
Автор

that seems pretty cool. i will try it! thank you!

legion_prex
Автор

For multiple layers, the first weights and biases are set randomly; how do you determine what to use for the next layer's weights and biases {when going forward, not backward)?

deeber