raflow 文档解析报错

Question

08:56:11 Page(4961): OCR started
08:56:29 Page(4961): OCR finished (17.76s)
08:56:31 Page(4961): Layout analysis (1.96s)
08:56:31 Page(4961): Table analysis (0.00s)
08:56:32 Page(4961): Text merged (0.00s)
08:56:36 Page(4961): Generate 10 chunks
08:56:39 Page(4961): Embedding chunks (3.18s)
08:56:41 Page(4961): Indexing done (1.77s). Task done (31.73s)
08:56:09 Task has been received.
08:56:12 Page(6173): OCR started
08:56:42 Page(6173): [ERROR]Internal server error while chunking: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Clip node. Name:p2o.Clip.6 Status Message: /onnxruntime_src/onnxruntime/core/framework/bfc_arena.cc:376 void* onnxruntime::BFCArena::AllocateRawInternal(size_t, bool, onnxruntime::Stream*, bool, onnxruntime::WaitNotificationFn) Available memory of 0 is smaller than requested bytes of 68290560
08:56:42 [ERROR][Exception]: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Clip node. Name:'p2o.Clip.6' Status Message: /onnxruntime_src/onnxruntime/core/framework/bfc_arena.cc:376 void* onnxruntime::BFCArena::AllocateRawInternal(size_t, bool, onnxruntime::Stream*, bool, onnxruntime::WaitNotificationFn) Available memory of 0 is smaller than requested bytes of 68290560
08:56:09 Task has been received.
08:56:12 Page(7385): OCR started
08:56:41 Page(7385): [ERROR]Internal server error while chunking: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Clip node. Name:p2o.Clip.6 Status Message: /onnxruntime_src/onnxruntime/core/framework/bfc_arena.cc:376 void* onnxruntime::BFCArena::AllocateRawInternal(size_t, bool, onnxruntime::Stream*, bool, onnxruntime::WaitNotificationFn) Available memory of 0 is smaller than requested bytes of 63184896
08:56:41 [ERROR][Exception]: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Clip node. Name:'p2o.Clip.6' Status Message: /onnxruntime_src/onnxruntime/core/framework/bfc_arena.cc:376 void* onnxruntime::BFCArena::AllocateRawInternal(size_t, bool, onnxruntime::Stream*, bool, onnxruntime::WaitNotificationFn) Available memory of 0 is smaller than requested bytes of 63184896
08:56:09 Task has been received.
08:56:12 Page(8597): OCR started
08:56:32 Page(8597): OCR finished (20.56s)
08:56:34 Page(8597): Layout analysis (1.33s)
08:56:34 Page(8597): Table analysis (0.00s)
08:56:34 Page(8597): Text merged (0.00s)
08:56:38 Page(8597): Generate 12 chunks
08:56:41 Page(8597): Embedding chunks (2.52s)
08:56:43 Page(8597): Indexing done (2.09s). Task done (33.70s)
08:56:10 Task has been received.
08:56:33 Page(157169): OCR started
08:56:54 Page(157169): OCR finished (21.32s)
08:56:56 Page(157169): Layout analysis (1.11s)
08:56:56 Page(157169): Table analysis (0.32s)
08:56:56 Page(157169): Text merged (0.00s)
08:57:00 Page(157169): Generate 16 chunks
08:57:04 Page(157169): Embedding chunks (3.66s)
08:57:06 Page(157169): Indexing done (1.85s). Task done (55.92s)
08:56:10 Task has been received.
08:56:35 Page(181193): OCR started
08:57:05 Page(181193): [ERROR]Internal server error while chunking: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Clip node. Name:p2o.Clip.6 Status Message: /onnxruntime_src/onnxruntime/core/framework/bfc_arena.cc:376 void* onnxruntime::BFCArena::AllocateRawInternal(size_t, bool, onnxruntime::Stream*, bool, onnxruntime::WaitNotificationFn) Available memory of 0 is smaller than requested bytes of 63897600
08:57:05 [ERROR][Exception]: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Clip node. onnxruntime onnxruntime::WaitNotificationFn) Available memory of 0 is smaller than requested bytes of 63897600
08:56:10 Task has been received.
08:56:43 Page(169181): OCR started
08:57:04 Page(169181): OCR finished (21.10s)
08:57:06 Page(169181): Layout analysis (1.87s)
08:57:06 Page(169181): Table analysis (0.00s)
08:57:06 Page(169181): Text merged (0.01s)
08:57:11 Page(169181): Generate 12 chunks
08:57:14 Page(169181): Embedding chunks (3.13s)
08:57:16 Page(169181): Indexing done (1.61s). Task done (65.54s)
08:56:10 Task has been received.
08:56:52 Page(109121): OCR started
08:57:12 Page(109121): OCR finished (19.53s)
08:57:13 Page(109~121): [ERROR]Internal server error while chunking: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Conv node. Name:/model.6/m.3/dilated_block/Conv Status Message: /onnxruntime_src/onnxruntime/core/framework/bfc_arena.cc:376 void* onnxruntime::BFCArena::AllocateRawInternal(size_t, bool, onnxruntime::Stream*, bool, onnxruntime::WaitNotificationFn) Available memory of 135266304 is smaller than requested bytes of 13674018573246399488
08:57:13 [ERROR][Exception]: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Conv node. Name:'/model.6/m.3/dilated_block/Conv' Status Message: /onnxruntime_src/onnxruntime/core/framework/bfc_arena.cc:376 void* onnxruntime::BFCArena::AllocateRawInternal(size_t, bool, onnxruntime::Stream*, bool, onnxruntime::WaitNotificationFn) Available memory of 135266304 is smaller than requested bytes of 13674018573246399488
08:56:11 Task has been received.

Answer 1

阿里嘎多学长整理AIGC生成，因移动端显示问题导致当前答案未能完全显示，请使用PC端查看更加详细的解答过程

raflow 文档解析报错

根据提供的信息，问题是 raflow 文档解析报错，错误信息中包含了 OCR 和 Layout 相关的信息。

可能的解决方案：

1. 检查 OCR 模型是否正确安装和配置，确保 OCR 模型能够正确地识别文本。
2. 检查 Layout 模型是否正确安装和配置，确保 Layout 模型能够正确地解析文档结构。
3. 检查文档是否正确地被读取和解析，确保文档的内容能够被正确地传递给 raflow。
4. 检查 raflow 的配置文件是否正确，确保 raflow 的配置能够正确地解析文档。

如果以上解决方案都不能解决问题，可以提供更多的错误信息和文档内容，以便更好地 debug 和解决问题。

核心代码：

import raflow

# 加载 OCR 模型
ocr_model = raflow.OcrModel('path/to/ocr/model')

# 加载 Layout 模型
layout_model = raflow.LayoutModel('path/to/layout/model')

# 读取文档
document = raflow.Document('path/to/document.pdf')

# 解析文档
document.parse(ocr_model, layout_model)

请注意，这只是一个示例代码，实际情况可能需要根据具体情况进行修改。