File size: 6,636 Bytes
231fb5f
a5e055b
231fb5f
7e2d83a
231fb5f
45cb2fc
7e2d83a
2796a5e
d6b2ea5
2796a5e
231fb5f
2796a5e
a5e055b
231fb5f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
fc8a89a
231fb5f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
fc8a89a
231fb5f
 
 
fc8a89a
231fb5f
 
 
 
fc8a89a
6f27c43
231fb5f
 
6f27c43
7877f6d
 
 
6f27c43
7877f6d
 
6f27c43
7877f6d
6f27c43
fc8a89a
231fb5f
 
 
 
 
 
fc8a89a
231fb5f
 
 
2b390ac
d6b2ea5
 
 
 
 
 
 
 
 
 
 
d3fde93
 
231fb5f
 
 
 
 
 
 
 
 
d3fde93
 
 
 
 
231fb5f
 
d3fde93
 
 
231fb5f
d3fde93
 
 
 
 
 
 
 
231fb5f
d3fde93
d6b2ea5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
231fb5f
 
 
 
 
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
import os
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, AutoProcessor, TextIteratorStreamer, BitsAndBytesConfig
import gradio as gr
from threading import Thread
import numpy as np
from PIL import Image
import subprocess
import spaces

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

# Constants
TITLE = "<h1><center>Phi 3.5 Multimodal (Text + Vision)</center></h1>"
DESCRIPTION = "# Phi-3.5 Multimodal Demo (Text + Vision)"

# Model configurations
TEXT_MODEL_ID = "microsoft/Phi-3.5-mini-instruct"
VISION_MODEL_ID = "microsoft/Phi-3.5-vision-instruct"

device = "cuda" if torch.cuda.is_available() else "cpu"

# Quantization config for text model
quantization_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_compute_dtype=torch.bfloat16,
    bnb_4bit_use_double_quant=True,
    bnb_4bit_quant_type="nf4"
)

# Load models and tokenizers
text_tokenizer = AutoTokenizer.from_pretrained(TEXT_MODEL_ID)
text_model = AutoModelForCausalLM.from_pretrained(
    TEXT_MODEL_ID,
    torch_dtype=torch.bfloat16,
    device_map="auto",
    quantization_config=quantization_config
)

vision_model = AutoModelForCausalLM.from_pretrained(
    VISION_MODEL_ID, 
    trust_remote_code=True, 
    torch_dtype="auto", 
    attn_implementation="flash_attention_2"
).to(device).eval()

vision_processor = AutoProcessor.from_pretrained(VISION_MODEL_ID, trust_remote_code=True)

# Helper functions
@spaces.GPU
def stream_text_chat(message, history, system_prompt, temperature=0.8, max_new_tokens=1024, top_p=1.0, top_k=20):
    conversation = [{"role": "system", "content": system_prompt}]
    for prompt, answer in history:
        conversation.extend([
            {"role": "user", "content": prompt},
            {"role": "assistant", "content": answer},
        ])
    conversation.append({"role": "user", "content": message})

    input_ids = text_tokenizer.apply_chat_template(conversation, add_generation_prompt=True, return_tensors="pt").to(text_model.device)
    streamer = TextIteratorStreamer(text_tokenizer, timeout=60.0, skip_prompt=True, skip_special_tokens=True)

    generate_kwargs = dict(
        input_ids=input_ids,
        max_new_tokens=max_new_tokens,
        do_sample=temperature > 0,
        top_p=top_p,
        top_k=top_k,
        temperature=temperature,
        eos_token_id=[128001, 128008, 128009],
        streamer=streamer,
    )

    with torch.no_grad():
        thread = Thread(target=text_model.generate, kwargs=generate_kwargs)
        thread.start()

    buffer = ""
    for new_text in streamer:
        buffer += new_text
        yield history + [[message, buffer]]

@spaces.GPU
def process_vision_query(image, text_input):
    prompt = f"<|user|>\n<|image_1|>\n{text_input}<|end|>\n<|assistant|>\n"
    
    # Ensure the image is in the correct format
    if isinstance(image, np.ndarray):
        # Convert numpy array to PIL Image
        image = Image.fromarray(image).convert("RGB")
    elif not isinstance(image, Image.Image):
        raise ValueError("Invalid image type. Expected PIL.Image.Image or numpy.ndarray")
    
    # Now process the image
    inputs = vision_processor(prompt, images=image, return_tensors="pt").to(device)
    
    with torch.no_grad():
        generate_ids = vision_model.generate(
            **inputs, 
            max_new_tokens=1000, 
            eos_token_id=vision_processor.tokenizer.eos_token_id
        )
    
    generate_ids = generate_ids[:, inputs['input_ids'].shape[1]:]
    response = vision_processor.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
    return response

# Combined chat function
def combined_chat(message, image, history, system_prompt, temperature, max_new_tokens, top_p, top_k):
    if image is not None:
        # Process image query
        response = process_vision_query(image, message)
        history.append((message, response))
        return history, None
    else:
        # Process text query
        return stream_text_chat(message, history, system_prompt, temperature, max_new_tokens, top_p, top_k), None

# Custom CSS
custom_css = """
body { background-color: #0b0f19; color: #e2e8f0; font-family: 'Arial', sans-serif;}
#custom-header { text-align: center; padding: 20px 0; background-color: #1a202c; margin-bottom: 20px; border-radius: 10px;}
#custom-header h1 { font-size: 2.5rem; margin-bottom: 0.5rem;}
#custom-header h1 .blue { color: #60a5fa;}
#custom-header h1 .pink { color: #f472b6;}
#custom-header h2 { font-size: 1.5rem; color: #94a3b8;}
.gradio-container { max-width: 100% !important;}
#component-0, #component-1, #component-2 { max-width: 100% !important;}
footer { text-align: center; margin-top: 2rem; color: #64748b;}
"""

# Custom HTML for the header
custom_header = """
<div id="custom-header">
    <h1><span class="blue">Phi 3.5</span> <span class="pink">Multimodal Assistant</span></h1>
    <h2>Text and Vision AI at Your Service</h2>
</div>
"""

# Gradio interface
with gr.Blocks(css=custom_css, theme=gr.themes.Base().set(
    body_background_fill="#0b0f19",
    body_text_color="#e2e8f0",
    button_primary_background_fill="#3b82f6",
    button_primary_background_fill_hover="#2563eb",
    button_primary_text_color="white",
    block_title_text_color="#94a3b8",
    block_label_text_color="#94a3b8",
)) as demo:
    gr.HTML(custom_header)

    chatbot = gr.Chatbot(height=400)
    msg = gr.Textbox(label="Message", placeholder="Type your message here...")
    image_input = gr.Image(label="Upload an Image (optional)", type="pil")

    with gr.Accordion("Advanced Options", open=False):
        system_prompt = gr.Textbox(value="You are a helpful assistant", label="System Prompt")
        temperature = gr.Slider(minimum=0, maximum=1, step=0.1, value=0.8, label="Temperature")
        max_new_tokens = gr.Slider(minimum=128, maximum=8192, step=1, value=1024, label="Max new tokens")
        top_p = gr.Slider(minimum=0.0, maximum=1.0, step=0.1, value=1.0, label="top_p")
        top_k = gr.Slider(minimum=1, maximum=20, step=1, value=20, label="top_k")
    
    submit_btn = gr.Button("Submit", variant="primary")
    clear_btn = gr.Button("Clear Chat", variant="secondary")

    submit_btn.click(combined_chat, [msg, image_input, chatbot, system_prompt, temperature, max_new_tokens, top_p, top_k], [chatbot, image_input])
    clear_btn.click(lambda: ([], None), None, [chatbot, image_input], queue=False)

    gr.HTML("<footer>Powered by Phi 3.5 Multimodal AI</footer>")

if __name__ == "__main__":
    demo.launch()