Skip to content

Conversation

@jurevreca12
Copy link
Collaborator

Adds a function del_initializer to ModelWrapper class. This function deletes an initializer by name.
This functionality is used in a chisel4ml transform that replaces quant nodes with custom nodes.
This PR should follow PR: #138

Fixed rounding_mode specifier in convert_quantized_bits
Commented out assertion on non-representability
…ay_and_auto_po2_scale

Merging changes from rounding_mode_new.
…nto HEAD

Merge changes from the new converter
…ed up by the normal converter flow.

Added an extra InferShape transform
@maltanar
Copy link
Collaborator

Thanks @jurevreca12 . Since the review for PR #138 is not yet done, I cherry-picked the key commit 3e793f2 here into #165, added a test and merged that.

@maltanar maltanar closed this Dec 20, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants