Add TextNet (#34979)
* WIP * Add config and modeling for Fast model * Refactor modeling and add tests * More changes * WIP * Add tests * Add conversion script * Add conversion scripts, integration tests, image processor * Fix style and copies * Add fast model to init * Add fast model in docs and other places * Fix import of cv2 * Rename image processing method * Fix build * Fix Build * fix style and fix copies * Fix build * Fix build * Fix Build * Clean up docstrings * Fix Build * Fix Build * Fix Build * Fix build * Add test for image_processing_fast and add documentation tests * some refactorings * Fix failing tests * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Introduce TextNet * Fix failures * Refactor textnet model * Fix failures * Add cv2 to setup * Fix failures * Fix failures * Add CV2 dependency * Fix bugs * Fix build issue * Fix failures * Remove textnet from modeling fast * Fix build and other things * Fix build * some cleanups * some cleanups * Some more cleanups * Fix build * Incorporate PR feedbacks * More cleanup * More cleanup * More cleanup * Fix build * Remove all the references of fast model * More cleanup * Fix build * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Fix Build * Fix build * Fix build * Fix build * Fix build * Fix build * Incorporate PR feedbacks * Fix style * Fix build * Incorporate PR feedbacks * Fix image processing mean and std * Incorporate PR feedbacks * fix build failure * Add assertion to image processor * Incorporate PR feedbacks * Incorporate PR feedbacks * fix style failures * fix build * Fix Imageclassification's linear layer, also introduce TextNetImageProcessor * Fix build * Fix build * Fix build * Fix build * Incorporate PR feedbacks * Incorporate PR feedbacks * Fix build * Incorporate PR feedbacks * Remove some script * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Incorporate PR feedbacks * Fix image processing in textnet * Incorporate PR Feedbacks * Fix CI failures * Fix failing test * Fix failing test * Fix failing test * Fix failing test * Fix failing test * Fix failing test * Add textnet to readme * Improve readability * Incorporate PR feedbacks * fix code style * fix key error and convert working * tvlt shouldn't be here * fix test modeling test * Fix tests, make fixup * Make fixup * Make fixup * Remove TEXTNET_PRETRAINED_MODEL_ARCHIVE_LIST * improve type annotation Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * Update tests/models/textnet/test_image_processing_textnet.py Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * improve type annotation Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * space typo Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * improve type annotation Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * Update src/transformers/models/textnet/configuration_textnet.py Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * make conv layer kernel sizes and strides default to None * Update src/transformers/models/textnet/modeling_textnet.py Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * Update src/transformers/models/textnet/modeling_textnet.py Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * fix keyword bug * add batch init and make fixup * Make fixup * Update integration test * Add figure * Update textnet.md * add testing and fix errors (classification, imgprocess) * fix error check * make fixup * make fixup * revert to original docstring * add make style * remove conflict for now * Update modeling_auto.py got a confusion in `timm_wrapper` - was giving some conflicts * Update tests/models/textnet/test_modeling_textnet.py Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * Update src/transformers/models/textnet/modeling_textnet.py Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * Update tests/models/textnet/test_modeling_textnet.py Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * Update src/transformers/models/textnet/modeling_textnet.py Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com> * add changes * Update textnet.md * add doc * add authors hf ckpt + rename * add feedback: classifier/docs --------- Co-authored-by: raghavanone <opensourcemaniacfreak@gmail.com> Co-authored-by: jadechoghari <jadechoghari@users.noreply.huggingface.co> Co-authored-by: Niels <niels.rogge1@gmail.com> Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com> Co-authored-by: Pavel Iakubovskii <qubvel@gmail.com>
This commit is contained in:
Jade Choghari
0
tests/models/textnet/__init__.py
Normal file
0
tests/models/textnet/__init__.py
Normal file
126
tests/models/textnet/test_image_processing_textnet.py
Normal file
126
tests/models/textnet/test_image_processing_textnet.py
Normal file
@@ -0,0 +1,126 @@
|
||||
# coding=utf-8
|
||||
# Copyright 2024 HuggingFace Inc.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
|
||||
|
||||
import unittest
|
||||
|
||||
from transformers.testing_utils import require_torch, require_vision
|
||||
from transformers.utils import is_vision_available
|
||||
|
||||
from ...test_image_processing_common import ImageProcessingTestMixin, prepare_image_inputs
|
||||
|
||||
|
||||
if is_vision_available():
|
||||
from transformers import TextNetImageProcessor
|
||||
|
||||
|
||||
class TextNetImageProcessingTester(unittest.TestCase):
|
||||
def __init__(
|
||||
self,
|
||||
parent,
|
||||
batch_size=7,
|
||||
num_channels=3,
|
||||
image_size=18,
|
||||
min_resolution=30,
|
||||
max_resolution=400,
|
||||
do_resize=True,
|
||||
size=None,
|
||||
size_divisor=32,
|
||||
do_center_crop=True,
|
||||
crop_size=None,
|
||||
do_normalize=True,
|
||||
image_mean=[0.48145466, 0.4578275, 0.40821073],
|
||||
image_std=[0.26862954, 0.26130258, 0.27577711],
|
||||
do_convert_rgb=True,
|
||||
):
|
||||
size = size if size is not None else {"shortest_edge": 20}
|
||||
crop_size = crop_size if crop_size is not None else {"height": 18, "width": 18}
|
||||
self.parent = parent
|
||||
self.batch_size = batch_size
|
||||
self.num_channels = num_channels
|
||||
self.image_size = image_size
|
||||
self.min_resolution = min_resolution
|
||||
self.max_resolution = max_resolution
|
||||
self.do_resize = do_resize
|
||||
self.size = size
|
||||
self.size_divisor = size_divisor
|
||||
self.do_center_crop = do_center_crop
|
||||
self.crop_size = crop_size
|
||||
self.do_normalize = do_normalize
|
||||
self.image_mean = image_mean
|
||||
self.image_std = image_std
|
||||
self.do_convert_rgb = do_convert_rgb
|
||||
|
||||
def prepare_image_processor_dict(self):
|
||||
return {
|
||||
"do_resize": self.do_resize,
|
||||
"size": self.size,
|
||||
"size_divisor": self.size_divisor,
|
||||
"do_center_crop": self.do_center_crop,
|
||||
"crop_size": self.crop_size,
|
||||
"do_normalize": self.do_normalize,
|
||||
"image_mean": self.image_mean,
|
||||
"image_std": self.image_std,
|
||||
"do_convert_rgb": self.do_convert_rgb,
|
||||
}
|
||||
|
||||
def expected_output_image_shape(self, images):
|
||||
return self.num_channels, self.crop_size["height"], self.crop_size["width"]
|
||||
|
||||
def prepare_image_inputs(self, equal_resolution=False, numpify=False, torchify=False):
|
||||
return prepare_image_inputs(
|
||||
batch_size=self.batch_size,
|
||||
num_channels=self.num_channels,
|
||||
min_resolution=self.min_resolution,
|
||||
max_resolution=self.max_resolution,
|
||||
equal_resolution=equal_resolution,
|
||||
numpify=numpify,
|
||||
torchify=torchify,
|
||||
)
|
||||
|
||||
|
||||
@require_torch
|
||||
@require_vision
|
||||
class TextNetImageProcessingTest(ImageProcessingTestMixin, unittest.TestCase):
|
||||
image_processing_class = TextNetImageProcessor if is_vision_available() else None
|
||||
|
||||
def setUp(self):
|
||||
super().setUp()
|
||||
self.image_processor_tester = TextNetImageProcessingTester(self)
|
||||
|
||||
@property
|
||||
def image_processor_dict(self):
|
||||
return self.image_processor_tester.prepare_image_processor_dict()
|
||||
|
||||
def test_image_processor_properties(self):
|
||||
image_processing = self.image_processing_class(**self.image_processor_dict)
|
||||
self.assertTrue(hasattr(image_processing, "do_resize"))
|
||||
self.assertTrue(hasattr(image_processing, "size"))
|
||||
self.assertTrue(hasattr(image_processing, "size_divisor"))
|
||||
self.assertTrue(hasattr(image_processing, "do_center_crop"))
|
||||
self.assertTrue(hasattr(image_processing, "center_crop"))
|
||||
self.assertTrue(hasattr(image_processing, "do_normalize"))
|
||||
self.assertTrue(hasattr(image_processing, "image_mean"))
|
||||
self.assertTrue(hasattr(image_processing, "image_std"))
|
||||
self.assertTrue(hasattr(image_processing, "do_convert_rgb"))
|
||||
|
||||
def test_image_processor_from_dict_with_kwargs(self):
|
||||
image_processor = self.image_processing_class.from_dict(self.image_processor_dict)
|
||||
self.assertEqual(image_processor.size, {"shortest_edge": 20})
|
||||
self.assertEqual(image_processor.crop_size, {"height": 18, "width": 18})
|
||||
|
||||
image_processor = self.image_processing_class.from_dict(self.image_processor_dict, size=42, crop_size=84)
|
||||
self.assertEqual(image_processor.size, {"shortest_edge": 42})
|
||||
self.assertEqual(image_processor.crop_size, {"height": 84, "width": 84})
|
||||
348
tests/models/textnet/test_modeling_textnet.py
Normal file
348
tests/models/textnet/test_modeling_textnet.py
Normal file
@@ -0,0 +1,348 @@
|
||||
# coding=utf-8
|
||||
# Copyright 2024 the Fast authors and The HuggingFace Inc. team. All rights reserved.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
"""Testing suite for the PyTorch TextNet model."""
|
||||
|
||||
import unittest
|
||||
|
||||
import requests
|
||||
from PIL import Image
|
||||
|
||||
from transformers import TextNetConfig
|
||||
from transformers.models.textnet.image_processing_textnet import TextNetImageProcessor
|
||||
from transformers.testing_utils import (
|
||||
require_torch,
|
||||
require_vision,
|
||||
slow,
|
||||
torch_device,
|
||||
)
|
||||
from transformers.utils import is_torch_available
|
||||
|
||||
from ...test_backbone_common import BackboneTesterMixin
|
||||
from ...test_configuration_common import ConfigTester
|
||||
from ...test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor
|
||||
from ...test_pipeline_mixin import PipelineTesterMixin
|
||||
|
||||
|
||||
if is_torch_available():
|
||||
import torch
|
||||
from torch import nn
|
||||
|
||||
from transformers import TextNetBackbone, TextNetForImageClassification, TextNetModel
|
||||
|
||||
|
||||
class TextNetConfigTester(ConfigTester):
|
||||
def create_and_test_config_common_properties(self):
|
||||
config = self.config_class(**self.inputs_dict)
|
||||
self.parent.assertTrue(hasattr(config, "hidden_sizes"))
|
||||
self.parent.assertTrue(hasattr(config, "num_attention_heads"))
|
||||
self.parent.assertTrue(hasattr(config, "num_encoder_blocks"))
|
||||
|
||||
|
||||
class TextNetModelTester:
|
||||
def __init__(
|
||||
self,
|
||||
parent,
|
||||
stem_kernel_size=3,
|
||||
stem_stride=2,
|
||||
stem_in_channels=3,
|
||||
stem_out_channels=32,
|
||||
stem_act_func="relu",
|
||||
dropout_rate=0,
|
||||
ops_order="weight_bn_act",
|
||||
conv_layer_kernel_sizes=[
|
||||
[[3, 3]],
|
||||
[[3, 3]],
|
||||
[[3, 3]],
|
||||
[[3, 3]],
|
||||
],
|
||||
conv_layer_strides=[
|
||||
[2],
|
||||
[2],
|
||||
[2],
|
||||
[2],
|
||||
],
|
||||
out_features=["stage1", "stage2", "stage3", "stage4"],
|
||||
out_indices=[1, 2, 3, 4],
|
||||
batch_size=3,
|
||||
num_channels=3,
|
||||
image_size=[32, 32],
|
||||
is_training=True,
|
||||
use_labels=True,
|
||||
num_labels=3,
|
||||
hidden_sizes=[32, 32, 32, 32, 32],
|
||||
):
|
||||
self.parent = parent
|
||||
self.stem_kernel_size = stem_kernel_size
|
||||
self.stem_stride = stem_stride
|
||||
self.stem_in_channels = stem_in_channels
|
||||
self.stem_out_channels = stem_out_channels
|
||||
self.act_func = stem_act_func
|
||||
self.dropout_rate = dropout_rate
|
||||
self.ops_order = ops_order
|
||||
self.conv_layer_kernel_sizes = conv_layer_kernel_sizes
|
||||
self.conv_layer_strides = conv_layer_strides
|
||||
|
||||
self.out_features = out_features
|
||||
self.out_indices = out_indices
|
||||
|
||||
self.batch_size = batch_size
|
||||
self.num_channels = num_channels
|
||||
self.image_size = image_size
|
||||
self.is_training = is_training
|
||||
self.use_labels = use_labels
|
||||
self.num_labels = num_labels
|
||||
self.hidden_sizes = hidden_sizes
|
||||
|
||||
self.num_stages = 5
|
||||
|
||||
def get_config(self):
|
||||
return TextNetConfig(
|
||||
stem_kernel_size=self.stem_kernel_size,
|
||||
stem_stride=self.stem_stride,
|
||||
stem_num_channels=self.stem_in_channels,
|
||||
stem_out_channels=self.stem_out_channels,
|
||||
act_func=self.act_func,
|
||||
dropout_rate=self.dropout_rate,
|
||||
ops_order=self.ops_order,
|
||||
conv_layer_kernel_sizes=self.conv_layer_kernel_sizes,
|
||||
conv_layer_strides=self.conv_layer_strides,
|
||||
out_features=self.out_features,
|
||||
out_indices=self.out_indices,
|
||||
hidden_sizes=self.hidden_sizes,
|
||||
image_size=self.image_size,
|
||||
)
|
||||
|
||||
def create_and_check_model(self, config, pixel_values, labels):
|
||||
model = TextNetModel(config=config)
|
||||
model.to(torch_device)
|
||||
model.eval()
|
||||
result = model(pixel_values)
|
||||
scale_h = self.image_size[0] // 32
|
||||
scale_w = self.image_size[1] // 32
|
||||
self.parent.assertEqual(
|
||||
result.last_hidden_state.shape,
|
||||
(self.batch_size, self.hidden_sizes[-1], scale_h, scale_w),
|
||||
)
|
||||
|
||||
def create_and_check_for_image_classification(self, config, pixel_values, labels):
|
||||
config.num_labels = self.num_labels
|
||||
model = TextNetForImageClassification(config)
|
||||
model.to(torch_device)
|
||||
model.eval()
|
||||
result = model(pixel_values, labels=labels)
|
||||
self.parent.assertEqual(result.logits.shape, (self.batch_size, self.num_labels))
|
||||
|
||||
def prepare_config_and_inputs(self):
|
||||
pixel_values = floats_tensor([self.batch_size, self.num_channels, self.image_size[0], self.image_size[1]])
|
||||
|
||||
labels = None
|
||||
if self.use_labels:
|
||||
labels = ids_tensor([self.batch_size], self.num_labels)
|
||||
|
||||
config = self.get_config()
|
||||
|
||||
return config, pixel_values, labels
|
||||
|
||||
def create_and_check_backbone(self, config, pixel_values, labels):
|
||||
model = TextNetBackbone(config=config)
|
||||
model.to(torch_device)
|
||||
model.eval()
|
||||
result = model(pixel_values)
|
||||
|
||||
# verify feature maps
|
||||
self.parent.assertEqual(len(result.feature_maps), len(config.out_features))
|
||||
scale_h = self.image_size[0] // 32
|
||||
scale_w = self.image_size[1] // 32
|
||||
self.parent.assertListEqual(
|
||||
list(result.feature_maps[0].shape), [self.batch_size, self.hidden_sizes[1], 8 * scale_h, 8 * scale_w]
|
||||
)
|
||||
|
||||
# verify channels
|
||||
self.parent.assertEqual(len(model.channels), len(config.out_features))
|
||||
self.parent.assertListEqual(model.channels, config.hidden_sizes[1:])
|
||||
|
||||
# verify backbone works with out_features=None
|
||||
config.out_features = None
|
||||
model = TextNetBackbone(config=config)
|
||||
model.to(torch_device)
|
||||
model.eval()
|
||||
result = model(pixel_values)
|
||||
|
||||
# verify feature maps
|
||||
self.parent.assertEqual(len(result.feature_maps), 1)
|
||||
scale_h = self.image_size[0] // 32
|
||||
scale_w = self.image_size[1] // 32
|
||||
self.parent.assertListEqual(
|
||||
list(result.feature_maps[0].shape), [self.batch_size, self.hidden_sizes[0], scale_h, scale_w]
|
||||
)
|
||||
|
||||
# verify channels
|
||||
self.parent.assertEqual(len(model.channels), 1)
|
||||
self.parent.assertListEqual(model.channels, [config.hidden_sizes[-1]])
|
||||
|
||||
def prepare_config_and_inputs_for_common(self):
|
||||
config_and_inputs = self.prepare_config_and_inputs()
|
||||
config, pixel_values, labels = config_and_inputs
|
||||
inputs_dict = {"pixel_values": pixel_values}
|
||||
return config, inputs_dict
|
||||
|
||||
|
||||
@require_torch
|
||||
class TextNetModelTest(ModelTesterMixin, PipelineTesterMixin, unittest.TestCase):
|
||||
"""
|
||||
Here we also overwrite some tests of test_modeling_common.py, as TextNet does not use input_ids, inputs_embeds,
|
||||
attention_mask and seq_length.
|
||||
"""
|
||||
|
||||
all_model_classes = (TextNetModel, TextNetForImageClassification, TextNetBackbone) if is_torch_available() else ()
|
||||
pipeline_model_mapping = (
|
||||
{"feature-extraction": TextNetModel, "image-classification": TextNetForImageClassification}
|
||||
if is_torch_available()
|
||||
else {}
|
||||
)
|
||||
|
||||
fx_compatible = False
|
||||
test_pruning = False
|
||||
test_resize_embeddings = False
|
||||
test_head_masking = False
|
||||
has_attentions = False
|
||||
|
||||
def setUp(self):
|
||||
self.model_tester = TextNetModelTester(self)
|
||||
self.config_tester = TextNetConfigTester(self, config_class=TextNetConfig, has_text_modality=False)
|
||||
|
||||
@unittest.skip(reason="TextNet does not output attentions")
|
||||
def test_attention_outputs(self):
|
||||
pass
|
||||
|
||||
@unittest.skip(reason="TextNet does not have input/output embeddings")
|
||||
def test_model_get_set_embeddings(self):
|
||||
pass
|
||||
|
||||
@unittest.skip(reason="TextNet does not use inputs_embeds")
|
||||
def test_inputs_embeds(self):
|
||||
pass
|
||||
|
||||
@unittest.skip(reason="TextNet does not support input and output embeddings")
|
||||
def test_model_common_attributes(self):
|
||||
pass
|
||||
|
||||
def test_model(self):
|
||||
config_and_inputs = self.model_tester.prepare_config_and_inputs()
|
||||
self.model_tester.create_and_check_model(*config_and_inputs)
|
||||
|
||||
def test_backbone(self):
|
||||
config_and_inputs = self.model_tester.prepare_config_and_inputs()
|
||||
self.model_tester.create_and_check_backbone(*config_and_inputs)
|
||||
|
||||
def test_initialization(self):
|
||||
config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
|
||||
|
||||
for model_class in self.all_model_classes:
|
||||
model = model_class(config=config)
|
||||
for name, module in model.named_modules():
|
||||
if isinstance(module, (nn.BatchNorm2d, nn.GroupNorm)):
|
||||
self.assertTrue(
|
||||
torch.all(module.weight == 1),
|
||||
msg=f"Parameter {name} of model {model_class} seems not properly initialized",
|
||||
)
|
||||
self.assertTrue(
|
||||
torch.all(module.bias == 0),
|
||||
msg=f"Parameter {name} of model {model_class} seems not properly initialized",
|
||||
)
|
||||
|
||||
def test_hidden_states_output(self):
|
||||
def check_hidden_states_output(inputs_dict, config, model_class):
|
||||
model = model_class(config)
|
||||
model.to(torch_device)
|
||||
model.eval()
|
||||
|
||||
with torch.no_grad():
|
||||
outputs = model(**self._prepare_for_class(inputs_dict, model_class))
|
||||
|
||||
hidden_states = outputs.encoder_hidden_states if config.is_encoder_decoder else outputs.hidden_states
|
||||
|
||||
self.assertEqual(len(hidden_states), self.model_tester.num_stages)
|
||||
|
||||
self.assertListEqual(
|
||||
list(hidden_states[0].shape[-2:]),
|
||||
[self.model_tester.image_size[0] // 2, self.model_tester.image_size[1] // 2],
|
||||
)
|
||||
|
||||
config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
|
||||
layers_type = ["preactivation", "bottleneck"]
|
||||
for model_class in self.all_model_classes:
|
||||
for layer_type in layers_type:
|
||||
config.layer_type = layer_type
|
||||
inputs_dict["output_hidden_states"] = True
|
||||
check_hidden_states_output(inputs_dict, config, model_class)
|
||||
|
||||
# check that output_hidden_states also work using config
|
||||
del inputs_dict["output_hidden_states"]
|
||||
config.output_hidden_states = True
|
||||
|
||||
check_hidden_states_output(inputs_dict, config, model_class)
|
||||
|
||||
@unittest.skip(reason="TextNet does not use feedforward chunking")
|
||||
def test_feed_forward_chunking(self):
|
||||
pass
|
||||
|
||||
def test_for_image_classification(self):
|
||||
config_and_inputs = self.model_tester.prepare_config_and_inputs()
|
||||
self.model_tester.create_and_check_for_image_classification(*config_and_inputs)
|
||||
|
||||
@slow
|
||||
def test_model_from_pretrained(self):
|
||||
model_name = "czczup/textnet-base"
|
||||
model = TextNetModel.from_pretrained(model_name)
|
||||
self.assertIsNotNone(model)
|
||||
|
||||
|
||||
@require_torch
|
||||
@require_vision
|
||||
class TextNetModelIntegrationTest(unittest.TestCase):
|
||||
@slow
|
||||
def test_inference_no_head(self):
|
||||
processor = TextNetImageProcessor.from_pretrained("czczup/textnet-base")
|
||||
model = TextNetModel.from_pretrained("czczup/textnet-base").to(torch_device)
|
||||
|
||||
# prepare image
|
||||
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
|
||||
image = Image.open(requests.get(url, stream=True).raw)
|
||||
inputs = processor(images=image, return_tensors="pt").to(torch_device)
|
||||
|
||||
# forward pass
|
||||
with torch.no_grad():
|
||||
output = model(**inputs)
|
||||
|
||||
# verify logits
|
||||
self.assertEqual(output.logits.shape, torch.Size([1, 2]))
|
||||
expected_slice_backbone = torch.tensor(
|
||||
[0.9210, 0.6099, 0.0000, 0.0000, 0.0000, 0.0000, 3.2207, 2.6602, 1.8925, 0.0000],
|
||||
device=torch_device,
|
||||
)
|
||||
self.assertTrue(torch.allclose(output.feature_maps[-1][0][10][12][:10], expected_slice_backbone, atol=1e-3))
|
||||
|
||||
|
||||
@require_torch
|
||||
# Copied from tests.models.bit.test_modeling_bit.BitBackboneTest with Bit->TextNet
|
||||
class TextNetBackboneTest(BackboneTesterMixin, unittest.TestCase):
|
||||
all_model_classes = (TextNetBackbone,) if is_torch_available() else ()
|
||||
config_class = TextNetConfig
|
||||
|
||||
has_attentions = False
|
||||
|
||||
def setUp(self):
|
||||
self.model_tester = TextNetModelTester(self)
|
||||
Reference in New Issue
Block a user