1
0

3 Revīzijas abedc4e5b6 ... de041fe413

Autors SHA1 Ziņojums Datums
  zhch158_admin de041fe413 feat(更新VL识别配置): 将GLM-OCR模块更改为PaddleOCR,更新相关参数以支持更大的图片尺寸和并发处理 1 nedēļu atpakaļ
  zhch158_admin 2fb3ee41cd feat(添加模型别名支持): 在paddle_local_daemon.sh中添加模型别名配置,并在启动和配置输出中显示 1 nedēļu atpakaļ
  zhch158_admin d1f002e9a7 feat(更新杨万益_福建农信文档配置): 修改输入输出路径和配置文件,更新日志文件路径 1 nedēļu atpakaļ

+ 5 - 0
ocr_tools/daemons/paddle_local_daemon.sh

@@ -25,6 +25,9 @@ HOST="0.0.0.0"
 MODEL_PATH="$HOME/models/paddleocr_vl/PaddlePaddle_PaddleOCR-VL-1.5-GGUF_PaddleOCR-VL-1.5.gguf"
 MMPROJ_PATH="$HOME/models/paddleocr_vl/PaddlePaddle_PaddleOCR-VL-1.5-GGUF_PaddleOCR-VL-1.5-mmproj.gguf"
 
+# 模型别名(对外暴露的模型 ID,对应 yaml 中的 model_name)
+MODEL_NAME="PaddleOCR-VL-1.5"
+
 # llama-server 参数
 CONTEXT_SIZE="16384"         # 上下文长度(需 >= max_tokens,推荐 8192-16384)
 GPU_LAYERS="99"              # Metal GPU 层数(99 表示全部)
@@ -93,6 +96,7 @@ start() {
     nohup llama-server \
         -m "$MODEL_PATH" \
         --mmproj "$MMPROJ_PATH" \
+        --alias $MODEL_NAME \
         --host $HOST \
         --port $PORT \
         --media-path $HOME/workspace \
@@ -202,6 +206,7 @@ config() {
     echo "  Conda 环境: $CONDA_ENV"
     echo "  Host: $HOST"
     echo "  Port: $PORT"
+    echo "  模型别名: $MODEL_NAME"
     echo "  主模型路径: $MODEL_PATH"
     echo "  多模态投影器: $MMPROJ_PATH"
     echo "  上下文长度: $CONTEXT_SIZE"

+ 11 - 7
ocr_tools/universal_doc_parser/config/bank_statement_paddle_vl_local.yaml

@@ -157,13 +157,17 @@ table_recognition_wired:
 # VL识别配置 - 使用 GLM-OCR(无线表格 + seal识别)
 # ============================================================
 vl_recognition:
-  module: "glmocr"
-  api_url: "http://localhost:8102/v1/chat/completions"
-  api_key: null  # 可选,如需要可填写
-  model: "glm-ocr"
-  max_image_size: 3500  # GLM-OCR 推荐的最大图片尺寸
-  resize_mode: 'max'    # 缩放模式: 'max' 保持宽高比, 'fixed' 固定尺寸
-  verify_ssl: false
+  module: "paddle"
+  backend: "http-client"
+  model_name: "PaddleOCR-VL-1.5"  # 与 paddle_local_daemon.sh 中 MODEL_NAME 一致
+  server_url: "http://localhost:8102"
+  max_image_size: 4096  # 🔧 添加:最大图片尺寸
+  resize_mode: 'max'    # 🔧 添加:缩放模式 ('max' 保持宽高比, 'fixed' 固定尺寸)
+  device: "cpu"
+  batch_size: 1
+  model_params:
+    max_concurrency: 10
+    http_timeout: 600
   
   # Task prompt mapping - 针对不同任务使用不同提示词
   task_prompt_mapping:

+ 7 - 4
ocr_tools/universal_doc_parser/main_v2.py

@@ -635,15 +635,18 @@ if __name__ == "__main__":
             # "input": "/Users/zhch158/workspace/data/流水分析/韩_中国银行图.pdf",
             # "output_dir": "/Users/zhch158/workspace/data/流水分析/韩_中国银行图/bank_statement_yusys_v4",
 
+            # "input": "/Users/zhch158/workspace/data/流水分析/杨万益_福建农信.pdf",
+            # "output_dir": "/Users/zhch158/workspace/data/流水分析/杨万益_福建农信/bank_statement_yusys_local",
+            # "config": "./config/bank_statement_yusys_local.yaml",
+
             "input": "/Users/zhch158/workspace/data/流水分析/杨万益_福建农信.pdf",
-            "output_dir": "/Users/zhch158/workspace/data/流水分析/杨万益_福建农信/bank_statement_yusys_local",
+            "output_dir": "/Users/zhch158/workspace/data/流水分析/杨万益_福建农信/bank_statement_paddle_vl_local",
+            "config": "./config/bank_statement_paddle_vl_local.yaml",
 
             # 日志文件
-            "log_file": "./output/logs/bank_statement_yusys_local/process.log",
+            "log_file": "./output/logs/bank_statement_paddle_vl_local/process.log",
 
             # 配置文件
-            "config": "./config/bank_statement_yusys_local.yaml",
-            # "config": "./config/bank_statement_paddle_vl_local.yaml",
             # "config": "./config/bank_statement_yusys_v4.yaml",
             # "config": "./config/bank_statement_yusys_v3.yaml",
             # "config": "./config/bank_statement_smart_router.yaml",