e96031413

import os
from pypdf import PdfReader
import re
from pathlib import Path

def get_pdf_title(pdf_path):
    """Get PDF title from metadata, fallback to content extraction if needed."""
    try:
        with open(pdf_path, 'rb') as file:
            pdf_reader = PdfReader(file)

e96031413

import requests
import os
from urllib.parse import urlparse

def convert_to_pdf_url(presentation_url):
    # Remove any trailing parameters or /edit
    base_url = presentation_url.split('/edit')[0].split('?')[0]
    # Add /export/pdf to the end
    return f"{base_url}/export/pdf"

e96031413

# git clone https://github.com/salesforce/BLIP
# cd BLIP
# touch BLIP_img2caption.py

import os
import json
from PIL import Image
import torch
from torchvision import transforms
from torchvision.transforms.functional import InterpolationMode

e96031413

import pandas as pd
from twstock import Stock
import argparse


def parse():
    parser = argparse.ArgumentParser()
    parser.add_argument(
        "--etf_code", type=str, default="00733",
    )

e96031413

from backtesting import Backtest, Strategy
from backtesting.lib import crossover

from FinMind.data import DataLoader
import pandas as pd

import talib
from talib import abstract

## 取得資料

e96031413

import whisper
import openai
import os

openai.api_key = "sk-XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX" 

# input audio file name
audio_file = r"C:\Users\User\Desktop\input.mp3"

# load the model and transcribe the audio

e96031413

import os
import cv2
import numpy as np
from tqdm import tqdm
import argparse

from facenet_pytorch import MTCNN

mtcnn = MTCNN(select_largest=True, min_face_size=64, post_process=False, device='cuda:0')

e96031413

import re
import os

urls = "https://drive.google.com/file/d/FILEID_1/view?usp=drive_link, https://drive.google.com/file/d/FILEID_2/view?usp=drive_link, https://drive.google.com/file/d/FILEID_3/view?usp=drive_link"
url_list = urls.split(', ')
pat = re.compile('https://drive.google.com/file/d/(.*)/view\?usp=drive_link')
for idx, url in enumerate(url_list):
    g = pat.match(url)
    id = g.group(1)
    down_url = f'https://drive.google.com/uc?id={id}'

e96031413

"""
https://github.com/d246810g2000/YOLOX/blob/main/datasets/train_val_data_split_coco.py
"""
import os
import cv2
import json
import random
import shutil
import xml.etree.ElementTree as ET
from tqdm import tqdm

e96031413

def resume_train(self, model):
    if self.args.resume:
        logger.info("resume training")
        if self.args.ckpt is None:
            ckpt_file = os.path.join(self.file_name, "latest" + "_ckpt.pth")
        else:
            ckpt_file = self.args.ckpt

        ckpt = torch.load(ckpt_file, map_location=self.device)
        # resume the model/optimizer state dict