sparrow-parse 0.3.3__tar.gz → 0.3.4__tar.gz

{sparrow-parse-0.3.3 → sparrow-parse-0.3.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sparrow-parse
-Version: 0.3.3
+Version: 0.3.4
 Summary: Sparrow Parse is a Python package for parsing and extracting information from documents.
 Home-page: https://github.com/katanaml/sparrow/tree/main/sparrow-data/parse
 Author: Andrej Baranovskij
@@ -87,6 +87,8 @@ Example:
 
 ## Parsing and extraction
 
+### HTML extractor
+
 ```
 from sparrow_parse.extractor.html_extractor import HTMLExtractor
 
@@ -119,6 +121,36 @@ Example:
 
 *debug* - `True`
 
+### Sparrow Parse VL (vision-language) extractor
+
+```
+extractor = VLLMExtractor()
+
+# export HF_TOKEN="hf_"
+config = {
+    "method": "huggingface",  # Could be 'huggingface' or 'local_gpu'
+    "hf_space": "katanaml/sparrow-qwen2-vl-7b",
+    "hf_token": os.getenv('HF_TOKEN'),
+    # Additional fields for local GPU inference
+    # "device": "cuda", "model_path": "model.pth"
+}
+
+# Use the factory to get the correct instance
+factory = InferenceFactory(config)
+model_inference_instance = factory.get_inference_instance()
+
+input_data = [
+    {
+        "image": "/Users/andrejb/Documents/work/epik/bankstatement/bonds_table.png",
+        "text_input": "retrieve financial instruments data. return response in JSON format"
+    }
+]
+
+# Now you can run inference without knowing which implementation is used
+result = extractor.run_inference(model_inference_instance, input_data, generic_query=False, debug=True)
+print("Inference Result:", result)
+```
+
 ## PDF optimization
 
 ```

{sparrow-parse-0.3.3 → sparrow-parse-0.3.4}/README.md

@@ -68,6 +68,8 @@ Example:
 
 ## Parsing and extraction
 
+### HTML extractor
+
 ```
 from sparrow_parse.extractor.html_extractor import HTMLExtractor
 
@@ -100,6 +102,36 @@ Example:
 
 *debug* - `True`
 
+### Sparrow Parse VL (vision-language) extractor
+
+```
+extractor = VLLMExtractor()
+
+# export HF_TOKEN="hf_"
+config = {
+    "method": "huggingface",  # Could be 'huggingface' or 'local_gpu'
+    "hf_space": "katanaml/sparrow-qwen2-vl-7b",
+    "hf_token": os.getenv('HF_TOKEN'),
+    # Additional fields for local GPU inference
+    # "device": "cuda", "model_path": "model.pth"
+}
+
+# Use the factory to get the correct instance
+factory = InferenceFactory(config)
+model_inference_instance = factory.get_inference_instance()
+
+input_data = [
+    {
+        "image": "/Users/andrejb/Documents/work/epik/bankstatement/bonds_table.png",
+        "text_input": "retrieve financial instruments data. return response in JSON format"
+    }
+]
+
+# Now you can run inference without knowing which implementation is used
+result = extractor.run_inference(model_inference_instance, input_data, generic_query=False, debug=True)
+print("Inference Result:", result)
+```
+
 ## PDF optimization
 
 ```

{sparrow-parse-0.3.3 → sparrow-parse-0.3.4}/setup.py

@@ -8,7 +8,7 @@ with open("requirements.txt", "r", encoding="utf-8") as fh:
 
 setup(
     name="sparrow-parse",
-    version="0.3.3",
+    version="0.3.4",
     author="Andrej Baranovskij",
     author_email="andrejus.baranovskis@gmail.com",
     description="Sparrow Parse is a Python package for parsing and extracting information from documents.",

sparrow-parse-0.3.4/sparrow_parse/__init__.py

@@ -0,0 +1 @@
+__version__ = '0.3.4'

{sparrow-parse-0.3.3 → sparrow-parse-0.3.4}/sparrow_parse.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sparrow-parse
-Version: 0.3.3
+Version: 0.3.4
 Summary: Sparrow Parse is a Python package for parsing and extracting information from documents.
 Home-page: https://github.com/katanaml/sparrow/tree/main/sparrow-data/parse
 Author: Andrej Baranovskij
@@ -87,6 +87,8 @@ Example:
 
 ## Parsing and extraction
 
+### HTML extractor
+
 ```
 from sparrow_parse.extractor.html_extractor import HTMLExtractor
 
@@ -119,6 +121,36 @@ Example:
 
 *debug* - `True`
 
+### Sparrow Parse VL (vision-language) extractor
+
+```
+extractor = VLLMExtractor()
+
+# export HF_TOKEN="hf_"
+config = {
+    "method": "huggingface",  # Could be 'huggingface' or 'local_gpu'
+    "hf_space": "katanaml/sparrow-qwen2-vl-7b",
+    "hf_token": os.getenv('HF_TOKEN'),
+    # Additional fields for local GPU inference
+    # "device": "cuda", "model_path": "model.pth"
+}
+
+# Use the factory to get the correct instance
+factory = InferenceFactory(config)
+model_inference_instance = factory.get_inference_instance()
+
+input_data = [
+    {
+        "image": "/Users/andrejb/Documents/work/epik/bankstatement/bonds_table.png",
+        "text_input": "retrieve financial instruments data. return response in JSON format"
+    }
+]
+
+# Now you can run inference without knowing which implementation is used
+result = extractor.run_inference(model_inference_instance, input_data, generic_query=False, debug=True)
+print("Inference Result:", result)
+```
+
 ## PDF optimization
 
 ```

sparrow-parse-0.3.3/sparrow_parse/__init__.py

@@ -1 +0,0 @@
-__version__ = '0.3.3'