Closed echarlaix closed 1 month ago
Fix generation for bloom architecture https://github.com/huggingface/optimum-intel/actions/runs/9232077090
Adapt _deduplicate_inputs to handle pkv when they are also instances of np.ndarray (which can happen since https://github.com/huggingface/optimum-intel/pull/727)
_deduplicate_inputs
np.ndarray
cc @eaidova
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.
Fix generation for bloom architecture https://github.com/huggingface/optimum-intel/actions/runs/9232077090
Adapt
_deduplicate_inputs
to handle pkv when they are also instances ofnp.ndarray
(which can happen since https://github.com/huggingface/optimum-intel/pull/727)cc @eaidova