Spaces:
Runtime error
Runtime error
# Based on https://github.com/haotian-liu/LLaVA. | |
import os | |
import json | |
import math | |
import torch | |
import argparse | |
from tqdm import tqdm | |
from decord import VideoReader, cpu | |
from llama_vstream.constants import IMAGE_TOKEN_INDEX, DEFAULT_IMAGE_TOKEN, DEFAULT_IM_START_TOKEN, DEFAULT_IM_END_TOKEN | |
from llama_vstream.conversation import conv_templates, SeparatorStyle | |
from llama_vstream.model.builder import load_pretrained_model | |
from llama_vstream.utils import disable_torch_init | |
from llama_vstream.mm_utils import tokenizer_image_token, get_model_name_from_path, KeywordsStoppingCriteria | |
def split_list(lst, n): | |
"""Split a list into n (roughly) equal-sized chunks""" | |
chunk_size = math.ceil(len(lst) / n) # integer division | |
return [lst[i:i+chunk_size] for i in range(0, len(lst), chunk_size)] | |
def get_chunk(lst, n, k): | |
chunks = split_list(lst, n) | |
return chunks[k] | |
def parse_args(): | |
""" | |
Parse command-line arguments. | |
""" | |
parser = argparse.ArgumentParser() | |
# Define the command-line arguments | |
parser.add_argument('--video_dir', help='Directory containing video files.', required=True) | |
parser.add_argument('--gt_file', help='Path to the ground truth file containing question.', required=True) | |
parser.add_argument('--output_dir', help='Directory to save the model results JSON.', required=True) | |
parser.add_argument('--output_name', help='Name of the file for storing results JSON.', required=True) | |
parser.add_argument("--model-path", type=str, default="facebook/opt-350m") | |
parser.add_argument("--model-base", type=str, default=None) | |
parser.add_argument("--conv-mode", type=str, default=None) | |
parser.add_argument("--num-chunks", type=int, default=1) | |
parser.add_argument("--chunk-idx", type=int, default=0) | |
parser.add_argument("--model-max-length", type=int, default=None) | |
return parser.parse_args() | |
def load_video(video_path): | |
vr = VideoReader(video_path, ctx=cpu(0)) | |
total_frame_num = len(vr) | |
fps = round(vr.get_avg_fps()) | |
frame_idx = [i for i in range(0, len(vr), fps)] | |
spare_frames = vr.get_batch(frame_idx).asnumpy() | |
return spare_frames | |
def run_inference(args): | |
""" | |
Run inference on ActivityNet QA DataSet using the Video-ChatGPT model. | |
Args: | |
args: Command-line arguments. | |
""" | |
# Initialize the model | |
model_name = get_model_name_from_path(args.model_path) | |
tokenizer, model, image_processor, context_len = load_pretrained_model(args.model_path, args.model_base, model_name, args.model_max_length) | |
# Load both ground truth file containing questions and answers | |
with open(args.gt_file) as file: | |
gt_questions = json.load(file) | |
gt_questions = get_chunk(gt_questions, args.num_chunks, args.chunk_idx) | |
# Create the output directory if it doesn't exist | |
if not os.path.exists(args.output_dir): | |
try: | |
os.makedirs(args.output_dir) | |
except Exception as e: | |
print(f'mkdir Except: {e}') | |
video_formats = ['.mp4', '.avi', '.mov', '.mkv'] | |
if args.num_chunks > 1: | |
output_name = f"{args.num_chunks}_{args.chunk_idx}" | |
else: | |
output_name = args.output_name | |
answers_file = os.path.join(args.output_dir, f"{output_name}.json") | |
ans_file = open(answers_file, "w") | |
for sample in tqdm(gt_questions, desc=f"cuda:{args.chunk_idx} "): | |
video_name = sample['video_id'] | |
question = sample['question'] | |
id = sample['id'] | |
answer = sample['answer'] | |
sample_set = {'id': id, 'question': question, 'answer': answer} | |
# Load the video file | |
for fmt in video_formats: # Added this line | |
temp_path = os.path.join(args.video_dir, f"{video_name}{fmt}") | |
if os.path.exists(temp_path): | |
video_path = temp_path | |
break | |
# Check if the video exists | |
if os.path.exists(video_path): | |
video = load_video(video_path) | |
video = image_processor.preprocess(video, return_tensors='pt')['pixel_values'].half().cuda() | |
video = [video] | |
qs = question | |
if model.config.mm_use_im_start_end: | |
qs = DEFAULT_IM_START_TOKEN + DEFAULT_IMAGE_TOKEN + DEFAULT_IM_END_TOKEN + '\n' + qs | |
else: | |
qs = DEFAULT_IMAGE_TOKEN + '\n' + qs | |
conv = conv_templates[args.conv_mode].copy() | |
conv.append_message(conv.roles[0], qs) | |
conv.append_message(conv.roles[1], None) | |
prompt = conv.get_prompt() | |
input_ids = tokenizer_image_token(prompt, tokenizer, IMAGE_TOKEN_INDEX, return_tensors='pt').unsqueeze(0).cuda() | |
stop_str = conv.sep if conv.sep_style != SeparatorStyle.TWO else conv.sep2 | |
keywords = [stop_str] | |
stopping_criteria = KeywordsStoppingCriteria(keywords, tokenizer, input_ids) | |
with torch.inference_mode(): | |
output_ids = model.generate( | |
input_ids, | |
images=video, | |
do_sample=True, | |
temperature=0.002, | |
max_new_tokens=1024, | |
use_cache=True, | |
stopping_criteria=[stopping_criteria]) | |
input_token_len = input_ids.shape[1] | |
n_diff_input_output = (input_ids != output_ids[:, :input_token_len]).sum().item() | |
if n_diff_input_output > 0: | |
print(f'[Warning] {n_diff_input_output} output_ids are not the same as the input_ids') | |
outputs = tokenizer.batch_decode(output_ids[:, input_token_len:], skip_special_tokens=True)[0] | |
outputs = outputs.strip() | |
if outputs.endswith(stop_str): | |
outputs = outputs[:-len(stop_str)] | |
outputs = outputs.strip() | |
sample_set['pred'] = outputs | |
ans_file.write(json.dumps(sample_set) + "\n") | |
ans_file.flush() | |
ans_file.close() | |
if __name__ == "__main__": | |
args = parse_args() | |
run_inference(args) | |