first commit

DeepSeek-OCR-master/DeepSeek-OCR-hf/run_dpsk_ocr.py

@@ -17,8 +17,8 @@ model = model.eval().cuda().to(torch.bfloat16)
 
 # prompt = "<image>\nFree OCR. "
 prompt = "<image>\n<|grounding|>Convert the document to markdown. "
-image_file = 'your_image.jpg'
-output_path = 'your/output/dir'
+image_file = '/workspace/2025-27484-M21472.pdf'
+output_path = '/workspace/output_hf'
 
 
 

DeepSeek-OCR-master/DeepSeek-OCR-vllm/config.py

@@ -21,8 +21,10 @@ MODEL_PATH = 'deepseek-ai/DeepSeek-OCR' # change to your model path
 # .jpg, .png, .jpeg: run_dpsk_ocr_image.py; 
 # Omnidocbench images path: run_dpsk_ocr_eval_batch.py
 
-INPUT_PATH = '' 
-OUTPUT_PATH = ''
+FILE_NAME='2025-27484-M21472.pdf'
+
+INPUT_PATH = f'/workspace/2025-27484-M21472.pdf' 
+OUTPUT_PATH = '/workspace/output/'
 
 PROMPT = '<image>\n<|grounding|>Convert the document to markdown.'
 # PROMPT = '<image>\nFree OCR.'

DeepSeek-OCR-master/DeepSeek-OCR-vllm/requirements.txt

@@ -0,0 +1,10 @@
+transformers
+tokenizers
+PyMuPDF
+img2pdf
+einops
+easydict
+addict 
+Pillow
+numpy
+matplotlib