15090929758

Committed 19 Oct 2024 05:10AM UTC coverage: 54.553%. First build

Build # 15090929758

Build Type

push

github

Committed by

web-flow

Commit Message

Merge pull request #228 from cosanlab/huggingface

WIP: Huggingface Integration

Run Details

702 of 1620 new or added lines in 46 files covered. (43.33%)

3409 of 6249 relevant lines covered (54.55%)

3.27 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

35.71

/feat/identity_detectors/facenet/facenet_test.py

from feat.utils import set_torch_device
from feat.identity_detectors.facenet.facenet_model import InceptionResnetV1


class Facenet:
    """Facenet Inception Resnet V1 model with optional loading of pretrained weights.

    Model parameters can be loaded based on pretraining on the VGGFace2 or CASIA-Webface
    datasets. Pretrained state_dicts are automatically downloaded on model instantiation if
    requested and cached in the torch cache. Subsequent instantiations use the cache rather than
    redownloading.

    Keyword Arguments:
        pretrained {str} -- Optional pretraining dataset. Either 'huggingface', 'vggface2' or 'casia-webface'.
            (default: {None})
        classify {bool} -- Whether the model should output classification probabilities or feature
            embeddings. (default: {False})
        num_classes {int} -- Number of output classes. If 'pretrained' is set and num_classes not
            equal to that used for the pretrained model, the final linear layer will be randomly
            initialized. (default: {None})
        dropout_prob {float} -- Dropout probability. (default: {0.6})
    """

    def __init__(
        self,
        pretrained="huggingface",
        classify=False,
        num_classes=None,
        dropout_prob=0.6,
        device="auto",
    ):
        super().__init__()

        if pretrained == "huggingface":
            self.model = InceptionResnetV1(
                pretrained=None,
                classify=classify,
                num_classes=num_classes,
                dropout_prob=dropout_prob,
                device=device,
            )
            self.model.from_pretrained("py-feat/facenet")
        else:
            pretrained = "vggface2"
            self.model = InceptionResnetV1(
                pretrained=pretrained,
                classify=classify,
                num_classes=num_classes,
                dropout_prob=dropout_prob,
                device=device,
            )

        self.device = set_torch_device(device)
        self.model.eval()

    def __call__(self, img):
        """
        img is of shape BxCxHxW --
        """

        return self.model.forward(img)

1	from feat.utils import set_torch_device	6✔
2	from feat.identity_detectors.facenet.facenet_model import InceptionResnetV1	6✔
3
4
5	class Facenet:	6✔
6	"""Facenet Inception Resnet V1 model with optional loading of pretrained weights.
7
8	Model parameters can be loaded based on pretraining on the VGGFace2 or CASIA-Webface
9	datasets. Pretrained state_dicts are automatically downloaded on model instantiation if
10	requested and cached in the torch cache. Subsequent instantiations use the cache rather than
11	redownloading.
12
13	Keyword Arguments:
14	pretrained {str} -- Optional pretraining dataset. Either 'huggingface', 'vggface2' or 'casia-webface'.
15	(default: {None})
16	classify {bool} -- Whether the model should output classification probabilities or feature
17	embeddings. (default: {False})
18	num_classes {int} -- Number of output classes. If 'pretrained' is set and num_classes not
19	equal to that used for the pretrained model, the final linear layer will be randomly
20	initialized. (default: {None})
21	dropout_prob {float} -- Dropout probability. (default: {0.6})
22	"""
23
24	def __init__(	6✔
25	self,
26	pretrained="huggingface",
27	classify=False,
28	num_classes=None,
29	dropout_prob=0.6,
30	device="auto",
31	):
32	super().__init__()	×
33
NEW 34	if pretrained == "huggingface":	×
NEW 35	self.model = InceptionResnetV1(	×
36	pretrained=None,
37	classify=classify,
38	num_classes=num_classes,
39	dropout_prob=dropout_prob,
40	device=device,
41	)
NEW 42	self.model.from_pretrained("py-feat/facenet")	×
43	else:
NEW 44	pretrained = "vggface2"	×
NEW 45	self.model = InceptionResnetV1(	×
46	pretrained=pretrained,
47	classify=classify,
48	num_classes=num_classes,
49	dropout_prob=dropout_prob,
50	device=device,
51	)
52
53	self.device = set_torch_device(device)	×
54	self.model.eval()	×
55
56	def __call__(self, img):	6✔
57	"""
58	img is of shape BxCxHxW --
59	"""
60
61	return self.model.forward(img)	×

cosanlab / py-feat / 15090929758

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous