122 lines
4.9 KiB
Python
122 lines
4.9 KiB
Python
import argparse
|
||
import os
|
||
import random
|
||
import numpy as np
|
||
import torch
|
||
import warnings
|
||
from PIL import Image
|
||
from transformers import AutoTokenizer, AutoModelForCausalLM
|
||
from model.model_videolm import MiniMindVideoLM
|
||
from model.VLMConfig import VLMConfig
|
||
from transformers import logging as hf_logging
|
||
from model.dataset import video2image
|
||
|
||
hf_logging.set_verbosity_error()
|
||
|
||
warnings.filterwarnings('ignore')
|
||
|
||
|
||
def count_parameters(model):
|
||
return sum(p.numel() for p in model.parameters() if p.requires_grad)
|
||
|
||
|
||
def init_model(lm_config, device):
|
||
tokenizer = AutoTokenizer.from_pretrained('./model/minimind_tokenizer')
|
||
if args.load == 0:
|
||
moe_path = '_moe' if args.use_moe else ''
|
||
modes = {0: 'pretrain_videolm', 1: 'sft_videolm'}
|
||
ckp = f'./{args.out_dir}/{modes[args.model_mode]}_{args.dim}{moe_path}.pth'
|
||
model = MiniMindVideoLM(lm_config)
|
||
state_dict = torch.load(ckp, map_location=device)
|
||
model.load_state_dict({k: v for k, v in state_dict.items() if 'mask' not in k}, strict=False)
|
||
else:
|
||
transformers_model_path = 'MiniMind2-V'
|
||
tokenizer = AutoTokenizer.from_pretrained(transformers_model_path)
|
||
model = AutoModelForCausalLM.from_pretrained(transformers_model_path, trust_remote_code=True)
|
||
|
||
print(f'VLM参数量：{sum(p.numel() for p in model.parameters() if p.requires_grad) / 1e6:.3f} 百万')
|
||
|
||
vision_model = MiniMindVideoLM.get_vision_model()
|
||
return model.eval().to(device), tokenizer, vision_model.eval().to(device)
|
||
|
||
|
||
def setup_seed(seed):
|
||
random.seed(seed)
|
||
np.random.seed(seed)
|
||
torch.manual_seed(seed)
|
||
torch.cuda.manual_seed(seed)
|
||
torch.cuda.manual_seed_all(seed)
|
||
torch.backends.cudnn.deterministic = True
|
||
torch.backends.cudnn.benchmark = False
|
||
|
||
|
||
if __name__ == "__main__":
|
||
parser = argparse.ArgumentParser(description="Chat with MiniMind")
|
||
parser.add_argument('--lora_name', default='None', type=str)
|
||
parser.add_argument('--out_dir', default='out', type=str)
|
||
parser.add_argument('--temperature', default=0.65, type=float)
|
||
parser.add_argument('--top_p', default=0.85, type=float)
|
||
parser.add_argument('--device', default='cuda' if torch.cuda.is_available() else 'cpu', type=str)
|
||
# MiniMind2-Small (26M)：(dim=512, n_layers=8)
|
||
# MiniMind2 (104M)：(dim=768, n_layers=16)
|
||
parser.add_argument('--dim', default=512, type=int)
|
||
parser.add_argument('--n_layers', default=8, type=int)
|
||
parser.add_argument('--max_seq_len', default=8192, type=int)
|
||
parser.add_argument('--use_moe', default=False, type=bool)
|
||
# 默认单图推理，设置为2为多图推理
|
||
parser.add_argument('--stream', default=True, type=bool)
|
||
parser.add_argument('--load', default=0, type=int, help="0: 原生torch权重，1: transformers加载")
|
||
parser.add_argument('--model_mode', default=0, type=int,
|
||
help="0: Pretrain模型，1: SFT模型")
|
||
args = parser.parse_args()
|
||
|
||
lm_config = VLMConfig(dim=args.dim, n_layers=args.n_layers, max_seq_len=args.max_seq_len, use_moe=args.use_moe)
|
||
|
||
model, tokenizer, vision_model = init_model(lm_config, args.device)
|
||
|
||
|
||
def chat_with_vlm(prompt, pixel_tensors, video_names):
|
||
messages = [{"role": "user", "content": prompt}]
|
||
|
||
new_prompt = tokenizer.apply_chat_template(
|
||
messages,
|
||
tokenize=False,
|
||
add_generation_prompt=True
|
||
)[-args.max_seq_len + 1:]
|
||
|
||
print(f'[Video]: {video_names}')
|
||
with torch.no_grad():
|
||
x = torch.tensor(tokenizer(new_prompt)['input_ids'], device=args.device).unsqueeze(0)
|
||
outputs = model.generate(
|
||
x,
|
||
eos_token_id=tokenizer.eos_token_id,
|
||
max_new_tokens=args.max_seq_len,
|
||
temperature=args.temperature,
|
||
top_p=args.top_p,
|
||
stream=True,
|
||
pad_token_id=tokenizer.pad_token_id,
|
||
pixel_tensors=pixel_tensors
|
||
)
|
||
print('🤖️: ', end='')
|
||
try:
|
||
if not args.stream:
|
||
print(tokenizer.decode(outputs.squeeze()[x.shape[1]:].tolist(), skip_special_tokens=True), end='')
|
||
else:
|
||
history_idx = 0
|
||
for y in outputs:
|
||
answer = tokenizer.decode(y[0].tolist(), skip_special_tokens=True)
|
||
if (answer and answer[-1] == '<EFBFBD>') or not answer:
|
||
continue
|
||
print(answer[history_idx:], end='', flush=True)
|
||
history_idx = len(answer)
|
||
except StopIteration:
|
||
print("No answer")
|
||
print('\n')
|
||
|
||
|
||
video_path = './dataset/eval_videos/video0.mp4'
|
||
prompt = f"what is a man driving down?\n{model.params.image_special_token}"
|
||
|
||
video_tensors = video2image(video_path).to(args.device).unsqueeze(0).unsqueeze(0)
|
||
chat_with_vlm(prompt, video_tensors, 'video0.mp4')
|