karpathy / nn-zero-to-hero

Neural Networks: Zero to Hero

karpathy/nn-zero-to-hero Issues

Fun solution to loop replacement (`makemore_part4_backprop.ipynb`)
Updated a month ago
BatchNorm1d in makemore part3 may have error
Closed 2 months ago
makemore_part4_backprop dhpreact exact part is False.
Updated 2 months ago
need Discussions :)
Closed 3 months ago
TypeError: unsupported operand type(s) for +: 'int' and 'Value'
Closed 4 months ago1
Error in makemore_part1_bigrams.ipynb
Updated 4 months ago1
Missing L.backward() before constructing graph
Closed 4 months ago1
zero_grad implementation correctness question
Closed a year ago3
Shouldn't names.txt be added to this repo?
Updated a year ago1
Factorized code for dC in makemore_part4_backprop.ipynb
Updated a year ago1
RuntimeError: Trying to backward through the graph a second time (or directly access saved tensors after they have already been freed). Saved intermediate values of the graph are freed when you call .backward() or autograd.grad(). Specify retain_graph=True if you need to backward through the graph a second time or if you need to access saved tensors after calling backward.
Updated a year ago1
Not bug, just curiosity
Updated a year ago3
Numerical instability in Google Colab - Part 4 of Makemore
Updated a year ago6
Discrepancy in makemore_part1_bigrams single sample draw
Closed a year ago2
Suggestion: Time Series Forecasting and Transfer Learning
Updated a year ago
Zero to hero
Updated a year ago
If I place more than 3 layers of FlattenConsecutive in makemore_part5_cnn1.ipynb, I get error: not enough values to unpack (expected 3, got 2)
Updated a year ago
difference on calculating loss for count-matrix
Updated a year ago
intro to neural networks first
Closed a year ago3
Not a real issue: Video links broken
Closed 2 years ago5
b
Updated 2 years ago1
Bug in makemore part-1, in computing the liklihood
Closed 2 years ago1