File size: 3,726 Bytes
baab6b5
 
 
 
 
 
 
 
 
 
 
 
 
 
106a698
baab6b5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d145f19
baab6b5
 
 
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
import io
import spaces
import argparse
import numpy as np
import torch
from decord import cpu, VideoReader, bridge
from transformers import AutoModelForCausalLM, AutoTokenizer
from transformers import BitsAndBytesConfig

MODEL_PATH = "THUDM/cogvlm2-llama3-caption"
DEVICE = 'cuda' if torch.cuda.is_available() else 'cpu'
TORCH_TYPE = torch.bfloat16 if torch.cuda.is_available() and torch.cuda.get_device_capability()[
    0] >= 8 else torch.float16

parser = argparse.ArgumentParser(description="Apollo tyre delay reasoning")
parser.add_argument('--quant', type=int, choices=[4, 8], help='Enable 4-bit or 8-bit precision loading', default=4)
args = parser.parse_args([])

def load_video(video_data, strategy='chat'):
    bridge.set_bridge('torch')
    mp4_stream = video_data
    num_frames = 24
    decord_vr = VideoReader(io.BytesIO(mp4_stream), ctx=cpu(0))
    frame_id_list = None
    total_frames = len(decord_vr)
    
    if strategy == 'base':
        clip_end_sec = 60
        clip_start_sec = 0
        start_frame = int(clip_start_sec * decord_vr.get_avg_fps())
        end_frame = min(total_frames,
                        int(clip_end_sec * decord_vr.get_avg_fps())) if clip_end_sec is not None else total_frames
        frame_id_list = np.linspace(start_frame, end_frame - 1, num_frames, dtype=int)
    elif strategy == 'chat':
        timestamps = decord_vr.get_frame_timestamp(np.arange(total_frames))
        timestamps = [i[0] for i in timestamps]
        max_second = round(max(timestamps)) + 1
        frame_id_list = []
        for second in range(max_second):
            closest_num = min(timestamps, key=lambda x: abs(x - second))
            index = timestamps.index(closest_num)
            frame_id_list.append(index)
            if len(frame_id_list) >= num_frames:
                break
                
    video_data = decord_vr.get_batch(frame_id_list)
    video_data = video_data.permute(3, 0, 1, 2)
    return video_data

# Configure quantization
quantization_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_compute_dtype=TORCH_TYPE,
    bnb_4bit_use_double_quant=True,
    bnb_4bit_quant_type="nf4"
)

tokenizer = AutoTokenizer.from_pretrained(
    MODEL_PATH,
    trust_remote_code=True,
)

model = AutoModelForCausalLM.from_pretrained(
    MODEL_PATH,
    torch_dtype=TORCH_TYPE,
    trust_remote_code=True,
    quantization_config=quantization_config,
    device_map="auto"
).eval()

@spaces.GPU
def predict(prompt, video_data, temperature):
    strategy = 'chat'
    video = load_video(video_data, strategy=strategy)
    history = []
    query = prompt
    inputs = model.build_conversation_input_ids(
        tokenizer=tokenizer,
        query=query,
        images=[video],
        history=history,
        template_version=strategy
    )
    
    inputs = {
        'input_ids': inputs['input_ids'].unsqueeze(0).to(DEVICE),
        'token_type_ids': inputs['token_type_ids'].unsqueeze(0).to(DEVICE),
        'attention_mask': inputs['attention_mask'].unsqueeze(0).to(DEVICE),
        'images': [[inputs['images'][0].to(DEVICE).to(TORCH_TYPE)]],
    }
    
    gen_kwargs = {
        "max_new_tokens": 2048,
        "pad_token_id": 128002,
        "top_k": 1,
        "do_sample": False,
        "top_p": 0.1,
        "temperature": temperature,
    }
    
    with torch.no_grad():
        outputs = model.generate(**inputs, **gen_kwargs)
        outputs = outputs[:, inputs['input_ids'].shape[1]:]
        response = tokenizer.decode(outputs[0], skip_special_tokens=True)
        return response

def inference(video, prompt):
    temperature = 0.8
    video_data = open(video, 'rb').read()
    response = predict(prompt, video_data, temperature)
    return response