mirror of
https://github.com/huggingface/transformers.git
synced 2025-07-13 01:30:04 +06:00
db72894b48
* fix * add tests * fix tests * Update tests/models/llava/test_processor_llava.py Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com> * fix * fix tests * update tests --------- Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
122 lines
4.7 KiB
Python
122 lines
4.7 KiB
Python
# Copyright 2021 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 json
|
|
import shutil
|
|
import tempfile
|
|
import unittest
|
|
|
|
from transformers import AutoProcessor, AutoTokenizer, LlamaTokenizerFast, LlavaProcessor
|
|
from transformers.testing_utils import require_torch, require_vision
|
|
from transformers.utils import is_vision_available
|
|
|
|
from ...test_processing_common import ProcessorTesterMixin
|
|
|
|
|
|
if is_vision_available():
|
|
from transformers import CLIPImageProcessor
|
|
|
|
|
|
@require_vision
|
|
class LlavaProcessorTest(ProcessorTesterMixin, unittest.TestCase):
|
|
processor_class = LlavaProcessor
|
|
|
|
def setUp(self):
|
|
self.tmpdirname = tempfile.mkdtemp()
|
|
|
|
image_processor = CLIPImageProcessor(do_center_crop=False)
|
|
tokenizer = LlamaTokenizerFast.from_pretrained("huggyllama/llama-7b")
|
|
processor_kwargs = self.prepare_processor_dict()
|
|
processor = LlavaProcessor(image_processor, tokenizer, **processor_kwargs)
|
|
processor.save_pretrained(self.tmpdirname)
|
|
|
|
def get_tokenizer(self, **kwargs):
|
|
return AutoProcessor.from_pretrained(self.tmpdirname, **kwargs).tokenizer
|
|
|
|
def get_image_processor(self, **kwargs):
|
|
return AutoProcessor.from_pretrained(self.tmpdirname, **kwargs).image_processor
|
|
|
|
def tearDown(self):
|
|
shutil.rmtree(self.tmpdirname)
|
|
|
|
def prepare_processor_dict(self):
|
|
return {"chat_template": "dummy_template"}
|
|
|
|
@unittest.skip(
|
|
"Skip because the model has no processor kwargs except for chat template and"
|
|
"chat template is saved as a separate file. Stop skipping this test when the processor"
|
|
"has new kwargs saved in config file."
|
|
)
|
|
def test_processor_to_json_string(self):
|
|
pass
|
|
|
|
def test_chat_template_is_saved(self):
|
|
processor_loaded = self.processor_class.from_pretrained(self.tmpdirname)
|
|
processor_dict_loaded = json.loads(processor_loaded.to_json_string())
|
|
# chat templates aren't serialized to json in processors
|
|
self.assertFalse("chat_template" in processor_dict_loaded.keys())
|
|
|
|
# they have to be saved as separate file and loaded back from that file
|
|
# so we check if the same template is loaded
|
|
processor_dict = self.prepare_processor_dict()
|
|
self.assertTrue(processor_loaded.chat_template == processor_dict.get("chat_template", None))
|
|
|
|
def test_can_load_various_tokenizers(self):
|
|
for checkpoint in ["Intel/llava-gemma-2b", "llava-hf/llava-1.5-7b-hf"]:
|
|
processor = LlavaProcessor.from_pretrained(checkpoint)
|
|
tokenizer = AutoTokenizer.from_pretrained(checkpoint)
|
|
self.assertEqual(processor.tokenizer.__class__, tokenizer.__class__)
|
|
|
|
def test_chat_template(self):
|
|
processor = LlavaProcessor.from_pretrained("llava-hf/llava-1.5-7b-hf")
|
|
expected_prompt = "USER: <image>\nWhat is shown in this image? ASSISTANT:"
|
|
|
|
messages = [
|
|
{
|
|
"role": "user",
|
|
"content": [
|
|
{"type": "image"},
|
|
{"type": "text", "text": "What is shown in this image?"},
|
|
],
|
|
},
|
|
]
|
|
|
|
formatted_prompt = processor.apply_chat_template(messages, add_generation_prompt=True)
|
|
self.assertEqual(expected_prompt, formatted_prompt)
|
|
|
|
@require_torch
|
|
@require_vision
|
|
def test_unstructured_kwargs_batched(self):
|
|
if "image_processor" not in self.processor_class.attributes:
|
|
self.skipTest(f"image_processor attribute not present in {self.processor_class}")
|
|
image_processor = self.get_component("image_processor")
|
|
tokenizer = self.get_component("tokenizer")
|
|
|
|
processor = self.processor_class(tokenizer=tokenizer, image_processor=image_processor)
|
|
self.skip_processor_without_typed_kwargs(processor)
|
|
|
|
input_str = ["lower newer", "upper older longer string"]
|
|
image_input = self.prepare_image_inputs() * 2
|
|
inputs = processor(
|
|
images=image_input,
|
|
text=input_str,
|
|
return_tensors="pt",
|
|
size={"height": 214, "width": 214},
|
|
padding="longest",
|
|
max_length=76,
|
|
)
|
|
|
|
self.assertEqual(inputs["pixel_values"].shape[2], 214)
|
|
|
|
self.assertEqual(len(inputs["input_ids"][0]), 5)
|