vlm4ocr 0.1.0__py3-none-any.whl → 0.3.0__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- vlm4ocr/__init__.py +3 -1
- vlm4ocr/assets/default_prompt_templates/ocr_JSON_system_prompt.txt +1 -0
- vlm4ocr/cli.py +276 -287
- vlm4ocr/data_types.py +109 -0
- vlm4ocr/ocr_engines.py +363 -195
- vlm4ocr/utils.py +386 -39
- vlm4ocr/vlm_engines.py +316 -190
- {vlm4ocr-0.1.0.dist-info → vlm4ocr-0.3.0.dist-info}/METADATA +5 -1
- vlm4ocr-0.3.0.dist-info/RECORD +17 -0
- vlm4ocr-0.1.0.dist-info/RECORD +0 -15
- {vlm4ocr-0.1.0.dist-info → vlm4ocr-0.3.0.dist-info}/WHEEL +0 -0
- {vlm4ocr-0.1.0.dist-info → vlm4ocr-0.3.0.dist-info}/entry_points.txt +0 -0
vlm4ocr/__init__.py
CHANGED
|
@@ -1,7 +1,9 @@
|
|
|
1
1
|
from .ocr_engines import OCREngine
|
|
2
|
-
from .vlm_engines import OllamaVLMEngine, OpenAIVLMEngine, AzureOpenAIVLMEngine
|
|
2
|
+
from .vlm_engines import BasicVLMConfig, OpenAIReasoningVLMConfig, OllamaVLMEngine, OpenAIVLMEngine, AzureOpenAIVLMEngine
|
|
3
3
|
|
|
4
4
|
__all__ = [
|
|
5
|
+
"BasicVLMConfig",
|
|
6
|
+
"OpenAIReasoningVLMConfig",
|
|
5
7
|
"OCREngine",
|
|
6
8
|
"OllamaVLMEngine",
|
|
7
9
|
"OpenAIVLMEngine",
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
You are a helpful assistant that can convert scanned documents into JSON format. Your output is accurate and well-formatted, starting with ```json and ending with ```. You will only output the JSON text without any additional explanations or comments. The JSON should include all text, tables, and lists with appropriate keys and values. You will ignore images, icons, or anything that can not be converted into text.
|