SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Allow passing arguments to model testers for CLIP-like models (#20044)

Browse Source 
* POC

* For more CLIP-like models

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
This commit is contained in:
Yih-Dar
2022-11-04 18:01:41 +01:00
committed by GitHub
parent 3bd0007e87
commit 2d02178e5c
5 changed files with 62 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
tests/models/clip/test_modeling_clip.py
View File

@@ -344,10 +344,16 @@ class CLIPTextModelTest(ModelTesterMixin, unittest.TestCase):
class CLIPModelTester: class CLIPModelTester:
def __init__(self, parent, is_training=True): def __init__(self, parent, text_kwargs=None, vision_kwargs=None, is_training=True):
if text_kwargs is None:
text_kwargs = {}
if vision_kwargs is None:
vision_kwargs = {}
self.parent = parent self.parent = parent
self.text_model_tester = CLIPTextModelTester(parent) self.text_model_tester = CLIPTextModelTester(parent, **text_kwargs)
self.vision_model_tester = CLIPVisionModelTester(parent) self.vision_model_tester = CLIPVisionModelTester(parent, **vision_kwargs)
self.is_training = is_training self.is_training = is_training
def prepare_config_and_inputs(self): def prepare_config_and_inputs(self):

22
tests/models/flava/test_modeling_flava.py
View File

@@ -746,17 +746,31 @@ class FlavaModelTester:
def __init__( def __init__(
self, self,
parent, parent,
text_kwargs=None,
image_kwargs=None,
multimodal_kwargs=None,
image_codebook_kwargs=None,
is_training=True, is_training=True,
hidden_size=32, hidden_size=32,
projection_dim=32, projection_dim=32,
initializer_range=0.02, initializer_range=0.02,
layer_norm_eps=1e-12, layer_norm_eps=1e-12,
): ):
if text_kwargs is None:
text_kwargs = {}
if image_kwargs is None:
image_kwargs = {}
if multimodal_kwargs is None:
multimodal_kwargs = {}
if image_codebook_kwargs is None:
image_codebook_kwargs = {}
self.parent = parent self.parent = parent
self.image_model_tester = FlavaImageModelTester(parent) self.image_model_tester = FlavaImageModelTester(parent, **image_kwargs)
self.text_model_tester = FlavaTextModelTester(parent) self.text_model_tester = FlavaTextModelTester(parent, **text_kwargs)
self.multimodal_model_tester = FlavaMultimodalModelTester(parent) self.multimodal_model_tester = FlavaMultimodalModelTester(parent, **multimodal_kwargs)
self.image_codebook_tester = FlavaImageCodebookTester(parent) self.image_codebook_tester = FlavaImageCodebookTester(parent, **image_codebook_kwargs)
self.is_training = is_training self.is_training = is_training
self.config_tester = ConfigTester(self, config_class=FlavaConfig, hidden_size=37) self.config_tester = ConfigTester(self, config_class=FlavaConfig, hidden_size=37)
self.hidden_size = hidden_size self.hidden_size = hidden_size

12
tests/models/groupvit/test_modeling_groupvit.py
View File

@@ -474,10 +474,16 @@ class GroupViTTextModelTest(ModelTesterMixin, unittest.TestCase):
class GroupViTModelTester: class GroupViTModelTester:
def __init__(self, parent, is_training=True): def __init__(self, parent, text_kwargs=None, vision_kwargs=None, is_training=True):
if text_kwargs is None:
text_kwargs = {}
if vision_kwargs is None:
vision_kwargs = {}
self.parent = parent self.parent = parent
self.text_model_tester = GroupViTTextModelTester(parent) self.text_model_tester = GroupViTTextModelTester(parent, **text_kwargs)
self.vision_model_tester = GroupViTVisionModelTester(parent) self.vision_model_tester = GroupViTVisionModelTester(parent, **vision_kwargs)
self.is_training = is_training self.is_training = is_training
def prepare_config_and_inputs(self): def prepare_config_and_inputs(self):

12
tests/models/owlvit/test_modeling_owlvit.py
View File

@@ -339,10 +339,16 @@ class OwlViTTextModelTest(ModelTesterMixin, unittest.TestCase):
class OwlViTModelTester: class OwlViTModelTester:
def __init__(self, parent, is_training=True): def __init__(self, parent, text_kwargs=None, vision_kwargs=None, is_training=True):
if text_kwargs is None:
text_kwargs = {}
if vision_kwargs is None:
vision_kwargs = {}
self.parent = parent self.parent = parent
self.text_model_tester = OwlViTTextModelTester(parent) self.text_model_tester = OwlViTTextModelTester(parent, **text_kwargs)
self.vision_model_tester = OwlViTVisionModelTester(parent) self.vision_model_tester = OwlViTVisionModelTester(parent, **vision_kwargs)
self.is_training = is_training self.is_training = is_training
self.text_config = self.text_model_tester.get_config().to_dict() self.text_config = self.text_model_tester.get_config().to_dict()
self.vision_config = self.vision_model_tester.get_config().to_dict() self.vision_config = self.vision_model_tester.get_config().to_dict()

20
tests/models/x_clip/test_modeling_x_clip.py
View File

@@ -436,12 +436,26 @@ class XCLIPTextModelTest(ModelTesterMixin, unittest.TestCase):
class XCLIPModelTester: class XCLIPModelTester:
def __init__(self, parent, projection_dim=64, mit_hidden_size=64, is_training=True): def __init__(
self,
parent,
text_kwargs=None,
vision_kwargs=None,
projection_dim=64,
mit_hidden_size=64,
is_training=True,
):
if text_kwargs is None:
text_kwargs = {}
if vision_kwargs is None:
vision_kwargs = {}
self.parent = parent self.parent = parent
self.projection_dim = projection_dim self.projection_dim = projection_dim
self.mit_hidden_size = mit_hidden_size self.mit_hidden_size = mit_hidden_size
self.text_model_tester = XCLIPTextModelTester(parent) self.text_model_tester = XCLIPTextModelTester(parent, **text_kwargs)
self.vision_model_tester = XCLIPVisionModelTester(parent) self.vision_model_tester = XCLIPVisionModelTester(parent, **vision_kwargs)
self.is_training = is_training self.is_training = is_training
def prepare_config_and_inputs(self): def prepare_config_and_inputs(self):