Vik Paruchuri commited on
Commit
478b8f8
·
1 Parent(s): b06e51f

Bump version

Browse files
marker/processors/debug.py CHANGED
@@ -93,6 +93,9 @@ class DebugProcessor(BaseProcessor):
93
  line_bboxes = []
94
  line_text = []
95
  for child in page.children:
 
 
 
96
  if child.block_type != BlockTypes.Line:
97
  continue
98
 
 
93
  line_bboxes = []
94
  line_text = []
95
  for child in page.children:
96
+ if child.removed:
97
+ continue
98
+
99
  if child.block_type != BlockTypes.Line:
100
  continue
101
 
marker/processors/llm/llm_image_description.py CHANGED
@@ -23,7 +23,7 @@ You will receive an image of a picture or figure. Your job will be to create a
23
  **Instructions:**
24
  1. Carefully examine the provided image.
25
  2. Analyze any text that was extracted from within the image.
26
- 3. Output a 3-4 sentence description of the image. Make sure there is enough specific detail to accurately describe the image. If there are numbers included, try to be specific.
27
  **Example:**
28
  Input:
29
  ```text
 
23
  **Instructions:**
24
  1. Carefully examine the provided image.
25
  2. Analyze any text that was extracted from within the image.
26
+ 3. Output a faithful description of the image. Make sure there is enough specific detail to accurately reconstruct the image. If the image is a figure or contains numeric data, include the numeric data in the output.
27
  **Example:**
28
  Input:
29
  ```text
pyproject.toml CHANGED
@@ -1,7 +1,7 @@
1
  [tool.poetry]
2
  name = "marker-pdf"
3
- version = "1.5.6"
4
- description = "Convert PDF to markdown with high speed and accuracy."
5
  authors = ["Vik Paruchuri <github@vikas.sh>"]
6
  readme = "README.md"
7
  license = "GPL-3.0-or-later"
 
1
  [tool.poetry]
2
  name = "marker-pdf"
3
+ version = "1.6.0"
4
+ description = "Convert documents to markdown with high speed and accuracy."
5
  authors = ["Vik Paruchuri <github@vikas.sh>"]
6
  readme = "README.md"
7
  license = "GPL-3.0-or-later"