File size: 12,035 Bytes
b381930 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 |
# Copyright 2022 The HuggingFace 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.
import unittest
import requests
from huggingface_hub import ImageToTextOutput
from transformers import MODEL_FOR_VISION_2_SEQ_MAPPING, TF_MODEL_FOR_VISION_2_SEQ_MAPPING, is_vision_available
from transformers.pipelines import ImageToTextPipeline, pipeline
from transformers.testing_utils import (
compare_pipeline_output_to_hub_spec,
is_pipeline_test,
require_tf,
require_torch,
require_vision,
slow,
)
from .test_pipelines_common import ANY
if is_vision_available():
from PIL import Image
else:
class Image:
@staticmethod
def open(*args, **kwargs):
pass
@is_pipeline_test
@require_vision
class ImageToTextPipelineTests(unittest.TestCase):
model_mapping = MODEL_FOR_VISION_2_SEQ_MAPPING
tf_model_mapping = TF_MODEL_FOR_VISION_2_SEQ_MAPPING
def get_test_pipeline(
self,
model,
tokenizer=None,
image_processor=None,
feature_extractor=None,
processor=None,
torch_dtype="float32",
):
pipe = ImageToTextPipeline(
model=model,
tokenizer=tokenizer,
feature_extractor=feature_extractor,
image_processor=image_processor,
processor=processor,
torch_dtype=torch_dtype,
)
examples = [
Image.open("./tests/fixtures/tests_samples/COCO/000000039769.png"),
"./tests/fixtures/tests_samples/COCO/000000039769.png",
]
return pipe, examples
def run_pipeline_test(self, pipe, examples):
outputs = pipe(examples)
self.assertEqual(
outputs,
[
[{"generated_text": ANY(str)}],
[{"generated_text": ANY(str)}],
],
)
@require_tf
def test_small_model_tf(self):
pipe = pipeline("image-to-text", model="hf-internal-testing/tiny-random-vit-gpt2", framework="tf")
image = "./tests/fixtures/tests_samples/COCO/000000039769.png"
outputs = pipe(image)
self.assertEqual(
outputs,
[
{
"generated_text": "growthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthGOGO"
},
],
)
outputs = pipe([image, image])
self.assertEqual(
outputs,
[
[
{
"generated_text": "growthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthGOGO"
}
],
[
{
"generated_text": "growthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthGOGO"
}
],
],
)
outputs = pipe(image, max_new_tokens=1)
self.assertEqual(
outputs,
[{"generated_text": "growth"}],
)
for single_output in outputs:
compare_pipeline_output_to_hub_spec(single_output, ImageToTextOutput)
@require_torch
def test_small_model_pt(self):
pipe = pipeline("image-to-text", model="hf-internal-testing/tiny-random-vit-gpt2")
image = "./tests/fixtures/tests_samples/COCO/000000039769.png"
outputs = pipe(image)
self.assertEqual(
outputs,
[
{
"generated_text": "growthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthGOGO"
},
],
)
outputs = pipe([image, image])
self.assertEqual(
outputs,
[
[
{
"generated_text": "growthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthGOGO"
}
],
[
{
"generated_text": "growthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthgrowthGOGO"
}
],
],
)
@require_torch
def test_small_model_pt_conditional(self):
pipe = pipeline("image-to-text", model="hf-internal-testing/tiny-random-BlipForConditionalGeneration")
image = "./tests/fixtures/tests_samples/COCO/000000039769.png"
prompt = "a photo of"
outputs = pipe(image, prompt=prompt)
self.assertTrue(outputs[0]["generated_text"].startswith(prompt))
@require_torch
def test_consistent_batching_behaviour(self):
pipe = pipeline("image-to-text", model="hf-internal-testing/tiny-random-BlipForConditionalGeneration")
image = "./tests/fixtures/tests_samples/COCO/000000039769.png"
prompt = "a photo of"
outputs = pipe([image, image], prompt=prompt)
self.assertTrue(outputs[0][0]["generated_text"].startswith(prompt))
self.assertTrue(outputs[1][0]["generated_text"].startswith(prompt))
outputs = pipe([image, image], prompt=prompt, batch_size=2)
self.assertTrue(outputs[0][0]["generated_text"].startswith(prompt))
self.assertTrue(outputs[1][0]["generated_text"].startswith(prompt))
from torch.utils.data import Dataset
class MyDataset(Dataset):
def __len__(self):
return 5
def __getitem__(self, i):
return "./tests/fixtures/tests_samples/COCO/000000039769.png"
dataset = MyDataset()
for batch_size in (1, 2, 4):
outputs = pipe(dataset, prompt=prompt, batch_size=batch_size if batch_size > 1 else None)
self.assertTrue(list(outputs)[0][0]["generated_text"].startswith(prompt))
self.assertTrue(list(outputs)[1][0]["generated_text"].startswith(prompt))
@slow
@require_torch
def test_large_model_pt(self):
pipe = pipeline("image-to-text", model="ydshieh/vit-gpt2-coco-en")
image = "./tests/fixtures/tests_samples/COCO/000000039769.png"
outputs = pipe(image)
self.assertEqual(outputs, [{"generated_text": "a cat laying on a blanket next to a cat laying on a bed "}])
outputs = pipe([image, image])
self.assertEqual(
outputs,
[
[{"generated_text": "a cat laying on a blanket next to a cat laying on a bed "}],
[{"generated_text": "a cat laying on a blanket next to a cat laying on a bed "}],
],
)
@slow
@require_torch
def test_generation_pt_blip(self):
pipe = pipeline("image-to-text", model="Salesforce/blip-image-captioning-base")
url = "https://huggingface.co/datasets/sayakpaul/sample-datasets/resolve/main/pokemon.png"
image = Image.open(requests.get(url, stream=True).raw)
outputs = pipe(image)
self.assertEqual(outputs, [{"generated_text": "a pink pokemon pokemon with a blue shirt and a blue shirt"}])
@slow
@require_torch
def test_generation_pt_git(self):
pipe = pipeline("image-to-text", model="microsoft/git-base-coco")
url = "https://huggingface.co/datasets/sayakpaul/sample-datasets/resolve/main/pokemon.png"
image = Image.open(requests.get(url, stream=True).raw)
outputs = pipe(image)
self.assertEqual(outputs, [{"generated_text": "a cartoon of a purple character."}])
@slow
@require_torch
def test_conditional_generation_pt_blip(self):
pipe = pipeline("image-to-text", model="Salesforce/blip-image-captioning-base")
url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg"
image = Image.open(requests.get(url, stream=True).raw)
prompt = "a photography of"
outputs = pipe(image, prompt=prompt)
self.assertEqual(outputs, [{"generated_text": "a photography of a volcano"}])
with self.assertRaises(ValueError):
outputs = pipe([image, image], prompt=[prompt, prompt])
@slow
@require_torch
def test_conditional_generation_pt_git(self):
pipe = pipeline("image-to-text", model="microsoft/git-base-coco")
url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg"
image = Image.open(requests.get(url, stream=True).raw)
prompt = "a photo of a"
outputs = pipe(image, prompt=prompt)
self.assertEqual(outputs, [{"generated_text": "a photo of a tent with a tent and a tent in the background."}])
with self.assertRaises(ValueError):
outputs = pipe([image, image], prompt=[prompt, prompt])
@slow
@require_torch
def test_conditional_generation_pt_pix2struct(self):
pipe = pipeline("image-to-text", model="google/pix2struct-ai2d-base")
url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg"
image = Image.open(requests.get(url, stream=True).raw)
prompt = "What does the label 15 represent? (1) lava (2) core (3) tunnel (4) ash cloud"
outputs = pipe(image, prompt=prompt)
self.assertEqual(outputs, [{"generated_text": "ash cloud"}])
with self.assertRaises(ValueError):
outputs = pipe([image, image], prompt=[prompt, prompt])
@slow
@require_tf
def test_large_model_tf(self):
pipe = pipeline("image-to-text", model="ydshieh/vit-gpt2-coco-en", framework="tf")
image = "./tests/fixtures/tests_samples/COCO/000000039769.png"
outputs = pipe(image)
self.assertEqual(outputs, [{"generated_text": "a cat laying on a blanket next to a cat laying on a bed "}])
outputs = pipe([image, image])
self.assertEqual(
outputs,
[
[{"generated_text": "a cat laying on a blanket next to a cat laying on a bed "}],
[{"generated_text": "a cat laying on a blanket next to a cat laying on a bed "}],
],
)
@slow
@require_torch
def test_conditional_generation_llava(self):
pipe = pipeline("image-to-text", model="llava-hf/bakLlava-v1-hf")
prompt = (
"<image>\nUSER: What does the label 15 represent? (1) lava (2) core (3) tunnel (4) ash cloud?\nASSISTANT:"
)
outputs = pipe(
"https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg",
prompt=prompt,
generate_kwargs={"max_new_tokens": 200},
)
self.assertEqual(
outputs,
[
{
"generated_text": "\nUSER: What does the label 15 represent? (1) lava (2) core (3) tunnel (4) ash cloud?\nASSISTANT: Lava"
}
],
)
@slow
@require_torch
def test_nougat(self):
pipe = pipeline("image-to-text", "facebook/nougat-base")
outputs = pipe("https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/nougat_paper.png")
self.assertEqual(
outputs,
[{"generated_text": "# Nougat: Neural Optical Understanding for Academic Documents\n\n Lukas Blec"}],
)
|