Victor writes here

Welcome — working notes on machine learning, from the math down to the metal.

Find me on

Posts

PyTorch Basics — dtype conversion, `.item()`, clamp, round, NLL loss, sigmoid, sqrt, in-place updates & `no_grad`, `requires_grad`, `torch.autograd.grad`, one-hot, (n,)↔(n,1), logical reductions (`any`/`all` + `dim`, axis-that-disappears, `keepdim`, dead-neuron fraction)
PyTorch nn Modules — `nn.Linear`, `nn.Dropout`, custom `nn.Module`, manual weight init, `nn.Parameter` (vs `register_buffer`), `kaiming_uniform_` (Kaiming vs Xavier), `state_dict`/`load_state_dict`
PyTorch Tensor Indexing: Slicing, Masking, Fancy Indexing — slicing (view) vs boolean masking (flattens) vs integer/fancy indexing; `gather`, `index_select`, `where`, `masked_fill`
Tensor Memory Layout: Storage, Strides, Contiguity, view/reshape/permute — storage/strides/`data_ptr`, contiguity, `view` vs `reshape`, `permute`/`transpose`, `.contiguous()`
Joining & Splitting Tensors: cat, stack, split, chunk — `cat` (existing dim) vs `stack` (new dim), `chunk` (count) vs `split` (size), `unbind`