einops
einops copied to clipboard
arogozhnikov
Flexible and powerful tensor operations for readable and reliable code (for pytorch, jax, TF and others)
As suggested on Twitter by @colah: https://twitter.com/ch402/status/1539774943214178304 > Is there an "einsum version" of gather()? Somehow gather is the most painful function to use when I need it (at least...
Great work! I discovered in https://github.com/arogozhnikov/einops/blob/master/einops/einops.py#L199 that you also support ellipsis. Its an important feature so you may want to add it to the documentation.
**Describe the bug** There appears to be several bugs when trying to use `einops` with TensorFlow tensors that have a dynamic shape. **Reproduction steps** Steps to reproduce the behavior: ```python...
The default value of `None` for the `bias_shape` parameter has an ambiguous meaning: either "inferred" or "no bias". So here's a clarification for that.
Hey! Loving `einops`, so much that now I feel a bit sad about standard `einsum` not being able to use descriptive names for dimensions. It would be amazing if `einops`...
Adds Flax Module's for einops operations. More details can be found in #153 einops supports JAX arrays, so I'm unsure if I should add tests or not ? CC: @arogozhnikov
**Describe the bug** When trying to call torchvision.models.feature_extraction.get_graph_node_names on a model that has an einops operation, the operation fail and gets the following error Traceback (most recent call last): File...
EinMix layer looks great but cannot be used with Jax since it's not a function. Would be great to have that EinMix layer in a Jax-based framework like Flax.
First of all: I like einops - readability in NN models is very much improved! a) Using einops as operation will create nearly unreadable graphs in tensorboard. To avoid this...
Metadata
Owner
Metadata
Flexible and powerful tensor operations for readable and reliable code (for pytorch, jax, TF and others)