mirror of
https://github.com/zenlm/vjepa2.git
synced 2026-10-30 19:33:29 +00:00
Co-authored-by: Nicolas Ballas <[email protected]> Co-authored-by: Adrien Bardes <[email protected]> Co-authored-by: David Fan <[email protected]> Co-authored-by: Quentin Garrido <[email protected]> Co-authored-by: Russell Howes <[email protected]> Co-authored-by: Mojtaba Komeili <[email protected]> Co-authored-by: Matthew Muckley <[email protected]> Co-authored-by: Koustuv Sinha <[email protected]> Co-authored-by: Artem Zholus <[email protected]>
94 lines
3.7 KiB
Python
94 lines
3.7 KiB
Python
# Copyright (c) Meta Platforms, Inc. and affiliates.
|
|
#
|
|
# This source code is licensed under the MIT license found in the
|
|
# LICENSE file in the root directory of this source tree.
|
|
|
|
import unittest
|
|
|
|
import torch
|
|
|
|
from src.models.vision_transformer import VIT_EMBED_DIMS, vit_tiny
|
|
|
|
|
|
class TestImageViT(unittest.TestCase):
|
|
def setUp(self) -> None:
|
|
self._vit_tiny = vit_tiny()
|
|
self.height, self.width = 224, 224
|
|
self.num_patches = (self.height // self._vit_tiny.patch_size) * (self.width // self._vit_tiny.patch_size)
|
|
|
|
def test_model_image_nomask_batchsize_4(self):
|
|
BS = 4
|
|
x = torch.rand((BS, 3, self.height, self.width))
|
|
y = self._vit_tiny(x)
|
|
self.assertIsInstance(y, torch.Tensor)
|
|
self.assertEqual(y.size(), (BS, self.num_patches, VIT_EMBED_DIMS["vit_tiny"]))
|
|
|
|
def test_model_image_nomask_batchsize_1(self):
|
|
BS = 1
|
|
x = torch.rand((BS, 3, self.height, self.width))
|
|
y = self._vit_tiny(x)
|
|
self.assertIsInstance(y, torch.Tensor)
|
|
self.assertEqual(y.size(), (BS, self.num_patches, VIT_EMBED_DIMS["vit_tiny"]))
|
|
|
|
def test_model_image_masked_batchsize_4(self):
|
|
BS = 4
|
|
mask_indices = [6, 7, 8]
|
|
masks = [torch.tensor(mask_indices, dtype=torch.int64) for _ in range(BS)]
|
|
x = torch.rand((BS, 3, self.height, self.width))
|
|
y = self._vit_tiny(x, masks=masks)
|
|
self.assertIsInstance(y, torch.Tensor)
|
|
self.assertEqual(y.size(), (BS, len(mask_indices), VIT_EMBED_DIMS["vit_tiny"]))
|
|
|
|
def test_model_image_masked_batchsize_1(self):
|
|
BS = 1
|
|
mask_indices = [6, 7, 8]
|
|
masks = [torch.tensor(mask_indices, dtype=torch.int64) for _ in range(BS)]
|
|
x = torch.rand((BS, 3, self.height, self.width))
|
|
y = self._vit_tiny(x, masks=masks)
|
|
self.assertIsInstance(y, torch.Tensor)
|
|
self.assertEqual(y.size(), (BS, len(mask_indices), VIT_EMBED_DIMS["vit_tiny"]))
|
|
|
|
|
|
class TestVideoViT(unittest.TestCase):
|
|
def setUp(self) -> None:
|
|
self.num_frames = 8
|
|
self._vit_tiny = vit_tiny(num_frames=8)
|
|
self.height, self.width = 224, 224
|
|
self.num_patches = (
|
|
(self.height // self._vit_tiny.patch_size)
|
|
* (self.width // self._vit_tiny.patch_size)
|
|
* (self.num_frames // self._vit_tiny.tubelet_size)
|
|
)
|
|
|
|
def test_model_video_nomask_batchsize_4(self):
|
|
BS = 4
|
|
x = torch.rand((BS, 3, self.num_frames, self.height, self.width))
|
|
y = self._vit_tiny(x)
|
|
self.assertIsInstance(y, torch.Tensor)
|
|
self.assertEqual(y.size(), (BS, self.num_patches, VIT_EMBED_DIMS["vit_tiny"]))
|
|
|
|
def test_model_video_nomask_batchsize_1(self):
|
|
BS = 1
|
|
x = torch.rand((BS, 3, self.num_frames, self.height, self.width))
|
|
y = self._vit_tiny(x)
|
|
self.assertIsInstance(y, torch.Tensor)
|
|
self.assertEqual(y.size(), (BS, self.num_patches, VIT_EMBED_DIMS["vit_tiny"]))
|
|
|
|
def test_model_video_masked_batchsize_4(self):
|
|
BS = 4
|
|
mask_indices = [6, 7, 8]
|
|
masks = [torch.tensor(mask_indices, dtype=torch.int64) for _ in range(BS)]
|
|
x = torch.rand((BS, 3, self.num_frames, self.height, self.width))
|
|
y = self._vit_tiny(x, masks=masks)
|
|
self.assertIsInstance(y, torch.Tensor)
|
|
self.assertEqual(y.size(), (BS, len(mask_indices), VIT_EMBED_DIMS["vit_tiny"]))
|
|
|
|
def test_model_video_masked_batchsize_1(self):
|
|
BS = 1
|
|
mask_indices = [6, 7, 8]
|
|
masks = [torch.tensor(mask_indices, dtype=torch.int64) for _ in range(BS)]
|
|
x = torch.rand((BS, 3, self.num_frames, self.height, self.width))
|
|
y = self._vit_tiny(x, masks=masks)
|
|
self.assertIsInstance(y, torch.Tensor)
|
|
self.assertEqual(y.size(), (BS, len(mask_indices), VIT_EMBED_DIMS["vit_tiny"]))
|