309afa20b7
* why does max_unpool2d feel slower than out.gradient ... * slightly cleaner * what happened to ruff * need to think about this some more * slightly faster now? * clean up, 1 more failing edge case * ok good * working TINY_BACKEND * nit doc wording * retry CI
1.4 KiB
Reduce
::: tinygrad.Tensor.sum ::: tinygrad.Tensor.prod ::: tinygrad.Tensor.max ::: tinygrad.Tensor.min ::: tinygrad.Tensor.any ::: tinygrad.Tensor.all ::: tinygrad.Tensor.isclose ::: tinygrad.Tensor.mean ::: tinygrad.Tensor.var ::: tinygrad.Tensor.std ::: tinygrad.Tensor.std_mean ::: tinygrad.Tensor.softmax ::: tinygrad.Tensor.log_softmax ::: tinygrad.Tensor.logsumexp ::: tinygrad.Tensor.logcumsumexp ::: tinygrad.Tensor.argmax ::: tinygrad.Tensor.argmin
Processing
::: tinygrad.Tensor.avg_pool2d ::: tinygrad.Tensor.max_pool2d ::: tinygrad.Tensor.max_unpool2d ::: tinygrad.Tensor.conv2d ::: tinygrad.Tensor.conv_transpose2d ::: tinygrad.Tensor.dot ::: tinygrad.Tensor.matmul ::: tinygrad.Tensor.einsum ::: tinygrad.Tensor.cumsum ::: tinygrad.Tensor.cummax ::: tinygrad.Tensor.triu ::: tinygrad.Tensor.tril ::: tinygrad.Tensor.interpolate ::: tinygrad.Tensor.scatter ::: tinygrad.Tensor.scatter_reduce ::: tinygrad.Tensor.sort ::: tinygrad.Tensor.topk
Neural Network (functional)
::: tinygrad.Tensor.linear ::: tinygrad.Tensor.sequential ::: tinygrad.Tensor.layernorm ::: tinygrad.Tensor.batchnorm ::: tinygrad.Tensor.dropout ::: tinygrad.Tensor.one_hot ::: tinygrad.Tensor.scaled_dot_product_attention ::: tinygrad.Tensor.binary_crossentropy ::: tinygrad.Tensor.binary_crossentropy_logits ::: tinygrad.Tensor.sparse_categorical_crossentropy ::: tinygrad.Tensor.cross_entropy ::: tinygrad.Tensor.nll_loss