chan/VLM-OCR
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Deepseek-OCR 환경 설정

Browse Source
This commit is contained in:
kyy
2025-10-27 13:50:19 +09:00
parent 153129649d
commit 96b47df624
21 changed files with 62 additions and 3481 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
model_services/deepseek_ocr/config.py
View File

@@ -8,25 +8,24 @@
BASE_SIZE = 1024
IMAGE_SIZE = 640
CROP_MODE = True
MIN_CROPS= 2
MAX_CROPS= 6 # max:9; If your GPU memory is small, it is recommended to set it to 6.
MAX_CONCURRENCY = 100 # If you have limited GPU memory, lower the concurrency count.
NUM_WORKERS = 64 # image pre-process (resize/padding) workers
MIN_CROPS = 2
MAX_CROPS = 6 # max:9; If your GPU memory is small, it is recommended to set it to 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' # change to your model path
MODEL_PATH = "deepseek-ai/DeepSeek-OCR" # change to your model path
# TODO: change INPUT_PATH
# .pdf: run_dpsk_ocr_pdf.py;
# .jpg, .png, .jpeg: run_dpsk_ocr_image.py;
# .pdf: run_dpsk_ocr_pdf.py;
# .jpg, .png, .jpeg: run_dpsk_ocr_image.py;
# Omnidocbench images path: run_dpsk_ocr_eval_batch.py
FILE_NAME='2025-27484-M21472.pdf'
INPUT_PATH = f'/workspace/2025-27484-M21472.pdf'
OUTPUT_PATH = '/workspace/output/'
INPUT_PATH = "/workspace/2018-0802140959-217049.pdf"
OUTPUT_PATH = "/workspace/output/"
PROMPT = '<image>\n<|grounding|>Convert the document to markdown.'
PROMPT = "<image>\n<|grounding|>Convert the document to markdown."
# PROMPT = '<image>\nFree OCR.'
# TODO commonly used prompts
# document: <image>\n<|grounding|>Convert the document to markdown.