Spaces:
Running
on
Zero
Running
on
Zero
File size: 2,789 Bytes
4f3b26b 161f68c 64036af 161f68c 64036af 161f68c 64036af 4f3b26b 161f68c 64036af 161f68c 64036af 161f68c 64036af 161f68c 64036af 161f68c |
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 |
import gradio as gr
import spaces
from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
from qwen_vl_utils import process_vision_info
import torch
import base64
from PIL import Image
from io import BytesIO
models = {
"Qwen/Qwen2-VL-7B-Instruct": Qwen2VLForConditionalGeneration.from_pretrained("Qwen/Qwen2-VL-7B-Instruct") #, torch_dtype="auto", device_map="auto")
}
processors = {
"Qwen/Qwen2-VL-7B-Instruct": AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct")
}
DESCRIPTION = "# Qwen2-VL Object Localization Demo"
def image_to_base64(image):
buffered = BytesIO()
image.save(buffered, format="PNG") # Save the image in memory as PNG
img_str = base64.b64encode(buffered.getvalue()).decode("utf-8") # Encode image to base64
return img_str
@spaces.GPU
def run_example(image, text_input, model_id="Qwen/Qwen2-VL-7B-Instruct"):
model = models[model_id].eval().cuda()
processor = processors[model_id]
messages = [
{
"role": "user",
"content": [
{"type": "image", "image": f"data:image;base64,{image_to_base64(image)}"},
{"type": "text", "text": f"Give a bounding box for {text_input}"},
],
}
]
text = processor.apply_chat_template(
messages, tokenize=False, add_generation_prompt=True
)
image_inputs, video_inputs = process_vision_info(messages)
inputs = processor(
text=[text],
images=image_inputs,
videos=video_inputs,
padding=True,
return_tensors="pt",
)
inputs = inputs.to("cuda")
generated_ids = model.generate(**inputs, max_new_tokens=128)
generated_ids_trimmed = [
out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]
output_text = processor.batch_decode(
generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
)
return output_text
css = """
#output {
height: 500px;
overflow: auto;
border: 1px solid #ccc;
}
"""
with gr.Blocks(css=css) as demo:
gr.Markdown(DESCRIPTION)
with gr.Tab(label="Qwen2-VL Input"):
with gr.Row():
with gr.Column():
input_img = gr.Image(label="Input Picture", type="pil")
model_selector = gr.Dropdown(choices=list(models.keys()), label="Model", value="Qwen/Qwen2-VL-7B-Instruct")
text_input = gr.Textbox(label="Description of Localization Target")
submit_btn = gr.Button(value="Submit")
with gr.Column():
output_text = gr.Textbox(label="Output Text")
submit_btn.click(run_example, [input_img, text_input, model_selector], [output_text])
demo.launch(debug=True) |