Add ctformer #191
Draft
Add ctformer #191
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
* convert to poetry * add make file * remove unused files * update license * throw warning if lambda layers exist * add custom onnx layer and hard sigmoid in it * connect hard sigmoid layer to converter * test hard sigmoid * add get config for onnx custom layer * send custom onnx objects map when loading from config * add keras-data-format-converter * add all interpolation modes to upsample * add all deeplab version tests * add convert_channels_first_to_last to testing
* raise on lamda layer exists * remove change_ordering from test and use our data format first to last with transform_inputs
* add OnnxReduceMean layer * replace lambda in new reduce mean layer * add get_special_data_format_handler for for reduce_mean_layer * update keras-data-format-converter
* move data format converter to dev dependecy * 0.0.34
* add special layer for erf operation * connect onnx operator to new tf layer
* convert all elemnt wise layers to use keras layer and fal back to operation instead of lambda * add sqrt operation layer * use keras merge layers only for two tensors * use oplambda for power
* multiple axes slicing suport * upgrade versions
* handling constants with a custom layer (mul) * modify mul to use lambda rather than custom layer * fixed eager tensor and dtype errors * added kwargs * gitignore + convert eager tensor to numpy * deleted custom layer * git ignore * added dtype conversion * elementwise sub with proper const handling * elementwise add const handling * up version * docstrings * padding * with padding fix * remove duplicated code * bugfix --------- Co-authored-by: ranhomri <ran.homri@tensorleap.ai>
* correct channels calculation when groups appear * update package version
* . * treat bollean in cons of shape * add cleaned name and elemntwise layers * up version * remove try catch * up version --------- Co-authored-by: ranhomri <ran.homri@tensorleap.ai> Co-authored-by: tensorleap <yotam@tensorleap.ai>
* . * added global max pool layer * pixellot o2k version * pixellot o2k version --------- Co-authored-by: tom Koren <tom.koren@tensorleap.ai> Co-authored-by: ranhomri <ran.homri@tensorleap.ai>
* . * . * . * . * . * update cache key
* fixes for yolov11 test * fixed test
* first version of working llama test with nference * refactored the code and removed unnecessary parts * removed unnecessary parts * fixing versions * fixing versions * FIXED THE CONDITION * added optimum * ignored test * ignored test
* fix reduce min for no axes cases * # This is a combination of 2 commits. # This is the 1st commit message: up o2k version # The commit message #2 will be skipped: # fixup! up o2k version * remove optimum temp * remove comment * up o2k version * change ubuntu * add torch version print * fix dinov2 test
* update to dev version and update convert_slice function * update to not dev version
* add mnist onnx test * fix conv layer to consider SAME LOWER AND UPPER params * update version
Co-authored-by: Idan <idan@idan.com>
Co-authored-by: Idan <idan@idan.com>
* . * . * > * > * > --------- Co-authored-by: Idan <idan@idan.com>
moving onnx to aws s3 changing default lstm batch to 1
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
10 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adding nn.Fold like module