#Boolean neural network with 8 bits weights during training

5 messages · Page 1 of 1 (latest)

tulip lark
#

this is my first serious project in c++ that ive been working on, id like to know where ive made mistakes or bad practices that should not be used in c++ in general, the project is unfinished but it works even as it is, im asking for opinions now because it would certainly be better if i'd have to rewrite just this thing instead of it+more https://github.com/logbyjungle/BitML
ignore the readme being terrible, i will make it better once its finished, its barely enough to explain how to use it
any suggestion will be appreciated

GitHub

a simple neural network header for c++ that uses only weights and biases with 1 bit values and uint8_t values during training - logbyjungle/BitML

harsh schooner
#

Nice work 🙂

tulip lark
harsh schooner
#

I haven’t looked at the implementation in-depth

#

Just a quick look 🙂