Rework add-new-model-like with modular and make test filenames coherent (#39612)
* remove tf/flax * fix * style * Update add_new_model_like.py * work in progress * continue * more cleanup * simplify and first final version * fixes -> it works * add linter checks * Update add_new_model_like.py * fix * add modular conversion at the end * Update add_new_model_like.py * add video processor * Update add_new_model_like.py * Update add_new_model_like.py * Update add_new_model_like.py * fix * Update image_processing_auto.py * Update image_processing_auto.py * fix post rebase * start test filenames replacement * rename all test_processor -> test_processing * fix copied from * add docstrings * Update add_new_model_like.py * fix regex * improve wording * Update add_new_model_like.py * Update add_new_model_like.py * Update add_new_model_like.py * start adding test * fix * fix * proper first test * tests * fix * fix * fix * fix * modular can be used from anywhere * protect import * fix * Update add_new_model_like.py * fix
This commit is contained in:
@@ -75,7 +75,7 @@ class ChameleonProcessorTest(ProcessorTesterMixin, unittest.TestCase):
|
||||
def prepare_processor_dict():
|
||||
return {"image_seq_length": 2} # fmt: skip
|
||||
|
||||
# Copied from tests.models.llava.test_processor_llava.LlavaProcessorTest.test_get_num_vision_tokens
|
||||
# Copied from tests.models.llava.test_processing_llava.LlavaProcessorTest.test_get_num_vision_tokens
|
||||
def test_get_num_vision_tokens(self):
|
||||
"Tests general functionality of the helper used internally in vLLM"
|
||||
|
||||
Reference in New Issue
Block a user