BASE_SIZE = 1024
IMAGE_SIZE = 768
CROP_MODE = True
MIN_CROPS= 2
MAX_CROPS= 6 # max:6
MAX_CONCURRENCY = 100 # If you have limited GPU memory, lower the concurrency count.
NUM_WORKERS = 64 # image pre-process (resize/padding) workers
PRINT_NUM_VIS_TOKENS = False
SKIP_REPEAT = True
MODEL_PATH = 'deepseek-ai/DeepSeek-OCR-2' # change to your model path

# TODO: change INPUT_PATH
# .pdf: run_dpsk_ocr_pdf.py;
# .jpg, .png, .jpeg: run_dpsk_ocr_image.py;
# Omnidocbench images path: run_dpsk_ocr_eval_batch.py


INPUT_PATH = 'doc/docstructbench_dianzishu_zhongwenzaixian-o.O-63686436.pdf_57.jpg'
OUTPUT_PATH = 'output/image/'

PROMPT = '<image>\n<|grounding|>Convert the document to markdown.'
# PROMPT = '<image>\nFree OCR.'
# PROMPT = '<image>\nParse the figure.'
# TODO commonly used prompts
# document: <image>\n<|grounding|>Convert the document to markdown.
# other image: <image>\n<|grounding|>OCR this image.
# without layouts: <image>\nFree OCR.
# figures in document: <image>\nParse the figure.
# general: <image>\nDescribe this image in detail.
# rec: <image>\nLocate <|ref|>xxxx<|/ref|> in the image.
# .......


from transformers import AutoTokenizer

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