Jackson
Jackson
Any movement here? Also experiencing this.
How does building wheels of packages with path dependencies work in this context?
I have been experiencing issues similar to those described above and it is due to mixing conda and pip installed packages inside a single conda env. Unfortunately mixing the two...
Or in the `extra` section of the `recipe.yaml` like `conda-lock` does.
Hey been looking at this project for a while, wanted to contribute. Seems like if you wanted to only throw an exception on lookup of a ambiguous axis name, instead...
Also, would you want to try to retain axes names on an outer product call?
Great, I'm adding this behavior to `contract_between`, `outer_product`, and `contract`. Do we also want to preserve remaining axis names after a call to `_contract_trace`?
It also appears that `_flatten_trace_edges` reshapes the node tensor, but does not regenerate the `axis_names` list leaving a mismatching number of names. This differs from `flatten_edges`. I assume this is...
There is serialization support that exports networks to json, with array formats stored in their backend specified format. Currently the only backend that currently implements the the `serialize/deserialize_tensor` methods is...
Hi @shun-lin you can extract the `weights` from a `fc_layer` in your `saved_model`, perform a decomposition and apply the properly decomposed weights to a `tn_layer` with the standard keras `set_weights`...