File size: 6,298 Bytes
e532cd7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
53b03cb
e532cd7
 
 
 
53b03cb
e532cd7
 
 
 
 
 
 
 
 
 
 
 
53b03cb
e532cd7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f53527d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e532cd7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
53b03cb
e532cd7
 
 
 
 
 
 
 
 
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
import gradio as gr
import spaces
from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
from qwen_vl_utils import process_vision_info
import torch
from PIL import Image
import subprocess
from datetime import datetime
import numpy as np
import os


# subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)

# models = {
#     "Qwen/Qwen2-VL-2B-Instruct": AutoModelForCausalLM.from_pretrained("Qwen/Qwen2-VL-2B-Instruct", trust_remote_code=True, torch_dtype="auto", _attn_implementation="flash_attention_2").cuda().eval()

# }
def array_to_image_path(image_array):
    # Convert numpy array to PIL Image
    img = Image.fromarray(np.uint8(image_array))
    
    # Generate a unique filename using timestamp
    timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
    filename = f"image_{timestamp}.png"
    
    # Save the image
    img.save(filename)
    
    # Get the full path of the saved image
    full_path = os.path.abspath(filename)
    
    return full_path
    
models = {
    "mateoluksenberg/Qwen-modelo-image": Qwen2VLForConditionalGeneration.from_pretrained("mateoluksenberg/Qwen-modelo-image", trust_remote_code=True, torch_dtype="auto").cuda().eval()

}

processors = {
    "mateoluksenberg/Qwen-modelo-image": AutoProcessor.from_pretrained("mateoluksenberg/Qwen-modelo-image", trust_remote_code=True)
}

DESCRIPTION = "[Qwen2-VL-2B Demo](https://huggingface.co/Qwen/Qwen2-VL-2B-Instruct)"

kwargs = {}
kwargs['torch_dtype'] = torch.bfloat16

user_prompt = '<|user|>\n'
assistant_prompt = '<|assistant|>\n'
prompt_suffix = "<|end|>\n"

@spaces.GPU
def run_example(image, text_input=None, model_id="mateoluksenberg/Qwen-modelo-image"):
    image_path = array_to_image_path(image)
    
    print(image_path)
    model = models[model_id]
    processor = processors[model_id]

    prompt = f"{user_prompt}<|image_1|>\n{text_input}{prompt_suffix}{assistant_prompt}"
    image = Image.fromarray(image).convert("RGB")
    messages = [
    {
            "role": "user",
            "content": [
                {
                    "type": "image",
                    "image": image_path,
                },
                {"type": "text", "text": text_input},
            ],
        }
    ]
    
    # Preparation for inference
    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")
    
    # Inference: Generation of the output
    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
    )

    "---------------"
    from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
    from qwen_vl_utils import process_vision_info
    
    # default: Load the model on the available device(s)
    model = Qwen2VLForConditionalGeneration.from_pretrained(
        "mateoluksenberg/Qwen-modelo-image", torch_dtype="auto", device_map="auto"
    )
    
    # We recommend enabling flash_attention_2 for better acceleration and memory saving, especially in multi-image and video scenarios.
    # model = Qwen2VLForConditionalGeneration.from_pretrained(
    #     "Qwen/Qwen2-VL-2B-Instruct",
    #     torch_dtype=torch.bfloat16,
    #     attn_implementation="flash_attention_2",
    #     device_map="auto",
    # )
    
    # default processer
    processor = AutoProcessor.from_pretrained("mateoluksenberg/Qwen-modelo-image")
    
    # The default range for the number of visual tokens per image in the model is 4-16384. You can set min_pixels and max_pixels according to your needs, such as a token count range of 256-1280, to balance speed and memory usage.
    # min_pixels = 256*28*28
    # max_pixels = 1280*28*28
    # processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-2B-Instruct", min_pixels=min_pixels, max_pixels=max_pixels)
    
    messages = [
        {
            "role": "user",
            "content": [
                {
                    "type": "image",
                    "image": image_path,
                },
                {"type": "text", "text": "Describe this image."},
            ],
        }
    ]
    
    # Preparation for inference
    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")
    
    # Inference: Generation of the output
    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
    )
    print(output_text)
    "---------------"
    
    return output_text[0]

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-2B Input"):
        with gr.Row():
            with gr.Column():
                input_img = gr.Image(label="Input Picture")
                model_selector = gr.Dropdown(choices=list(models.keys()), label="Model", value="mateoluksenberg/Qwen-modelo-image")
                text_input = gr.Textbox(label="Question")
                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.queue(api_open=False)
demo.launch(debug=True)