sparrow-parse 0.3.6__tar.gz → 0.3.8__tar.gz

{sparrow-parse-0.3.6 → sparrow-parse-0.3.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sparrow-parse
-Version: 0.3.6
+Version: 0.3.8
 Summary: Sparrow Parse is a Python package (part of Sparrow) for parsing and extracting information from documents.
 Home-page: https://github.com/katanaml/sparrow/tree/main/sparrow-data/parse
 Author: Andrej Baranovskij
@@ -65,7 +65,7 @@ input_data = [
 results_array, num_pages = extractor.run_inference(model_inference_instance, input_data, generic_query=False,
                                  debug_dir="/data/",
                                  debug=True,
-                                 mode="static")
+                                 mode=None)
 
 for i, result in enumerate(results_array):
     print(f"Result for page {i + 1}:", result)
@@ -74,6 +74,10 @@ print(f"Number of pages: {num_pages}")
 
 Use `mode="static"` if you want to simulate LLM call, without executing LLM backend.
 
+Method `run_inference` will return results and number of pages processed.
+
+Note: GPU backend `katanaml/sparrow-qwen2-vl-7b` is private, to be able to run below command, you need to create your own backend on Hugging Face space using [code](https://github.com/katanaml/sparrow/tree/main/sparrow-data/parse/sparrow_parse/vllm/infra/qwen2_vl_7b) from Sparrow Parse.
+
 ## PDF pre-processing
 
 ```

{sparrow-parse-0.3.6 → sparrow-parse-0.3.8}/README.md

@@ -46,7 +46,7 @@ input_data = [
 results_array, num_pages = extractor.run_inference(model_inference_instance, input_data, generic_query=False,
                                  debug_dir="/data/",
                                  debug=True,
-                                 mode="static")
+                                 mode=None)
 
 for i, result in enumerate(results_array):
     print(f"Result for page {i + 1}:", result)
@@ -55,6 +55,10 @@ print(f"Number of pages: {num_pages}")
 
 Use `mode="static"` if you want to simulate LLM call, without executing LLM backend.
 
+Method `run_inference` will return results and number of pages processed.
+
+Note: GPU backend `katanaml/sparrow-qwen2-vl-7b` is private, to be able to run below command, you need to create your own backend on Hugging Face space using [code](https://github.com/katanaml/sparrow/tree/main/sparrow-data/parse/sparrow_parse/vllm/infra/qwen2_vl_7b) from Sparrow Parse.
+
 ## PDF pre-processing
 
 ```

{sparrow-parse-0.3.6 → sparrow-parse-0.3.8}/setup.py

@@ -8,7 +8,7 @@ with open("requirements.txt", "r", encoding="utf-8") as fh:
 
 setup(
     name="sparrow-parse",
-    version="0.3.6",
+    version="0.3.8",
     author="Andrej Baranovskij",
     author_email="andrejus.baranovskis@gmail.com",
     description="Sparrow Parse is a Python package (part of Sparrow) for parsing and extracting information from documents.",

sparrow-parse-0.3.8/sparrow_parse/__init__.py

@@ -0,0 +1 @@
+__version__ = '0.3.8'

{sparrow-parse-0.3.6 → sparrow-parse-0.3.8}/sparrow_parse/extractors/vllm_extractor.py

@@ -30,21 +30,16 @@ class VLLMExtractor(object):
                                                                                  debug_dir,
                                                                                  True)
 
-            # Run inference on each page
-            for page_num, output_file in enumerate(output_files):
-                input_data[0]["file_path"] = output_file
-                if debug:
-                    print(f"Running inference on page {page_num + 1}...")
+            input_data[0]["file_path"] = output_files
 
-                # Run inference on the page
-                result = model_inference_instance.inference(input_data, mode)
-                results_array.append(result)
+            # Run inference on the page
+            results_array = model_inference_instance.inference(input_data, mode)
 
             shutil.rmtree(temp_dir, ignore_errors=True)
             return results_array, num_pages
 
-        result = model_inference_instance.inference(input_data)
-        results_array.append(result)
+        input_data[0]["file_path"] = [input_data[0]["file_path"]]
+        results_array = model_inference_instance.inference(input_data)
 
         return results_array, 1
 
@@ -80,7 +75,7 @@ if __name__ == "__main__":
     # results_array, num_pages = extractor.run_inference(model_inference_instance, input_data, generic_query=False,
     #                                  debug_dir="/Users/andrejb/infra/shared/katana-git/sparrow/sparrow-ml/llm/data/",
     #                                  debug=True,
-    #                                  mode="static")
+    #                                  mode=None)
     #
     # for i, result in enumerate(results_array):
     #     print(f"Result for page {i + 1}:", result)

sparrow-parse-0.3.8/sparrow_parse/vllm/huggingface_inference.py

@@ -0,0 +1,60 @@
+from gradio_client import Client, handle_file
+from sparrow_parse.vllm.inference_base import ModelInference
+import json
+import os
+import ast
+
+
+class HuggingFaceInference(ModelInference):
+    def __init__(self, hf_space, hf_token):
+        self.hf_space = hf_space
+        self.hf_token = hf_token
+
+
+    def process_response(self, output_text):
+        json_string = output_text
+
+        json_string = json_string.strip("[]'")
+        json_string = json_string.replace("```json\n", "").replace("\n```", "")
+        json_string = json_string.replace("'", "")
+
+        try:
+            formatted_json = json.loads(json_string)
+            return json.dumps(formatted_json, indent=2)
+        except json.JSONDecodeError as e:
+            print("Failed to parse JSON:", e)
+            return output_text
+
+
+    def inference(self, input_data, mode=None):
+        if mode == "static":
+            simple_json = self.get_simple_json()
+            return [simple_json]
+
+        client = Client(self.hf_space, hf_token=self.hf_token)
+
+        # Extract and prepare the absolute paths for all file paths in input_data
+        file_paths = [
+            os.path.abspath(file_path)
+            for data in input_data
+            for file_path in data["file_path"]
+        ]
+
+        # Validate file existence and prepare files for the Gradio client
+        image_files = [handle_file(path) for path in file_paths if os.path.exists(path)]
+
+        results = client.predict(
+            input_imgs=image_files,
+            text_input=input_data[0]["text_input"],  # Single shared text input for all images
+            api_name="/run_inference"  # Specify the Gradio API endpoint
+        )
+
+        # Convert the string into a Python list
+        parsed_results = ast.literal_eval(results)
+
+        results_array = []
+        for page_output in parsed_results:
+            page_result = self.process_response(page_output)
+            results_array.append(page_result)
+
+        return results_array

{sparrow-parse-0.3.6 → sparrow-parse-0.3.8}/sparrow_parse.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sparrow-parse
-Version: 0.3.6
+Version: 0.3.8
 Summary: Sparrow Parse is a Python package (part of Sparrow) for parsing and extracting information from documents.
 Home-page: https://github.com/katanaml/sparrow/tree/main/sparrow-data/parse
 Author: Andrej Baranovskij
@@ -65,7 +65,7 @@ input_data = [
 results_array, num_pages = extractor.run_inference(model_inference_instance, input_data, generic_query=False,
                                  debug_dir="/data/",
                                  debug=True,
-                                 mode="static")
+                                 mode=None)
 
 for i, result in enumerate(results_array):
     print(f"Result for page {i + 1}:", result)
@@ -74,6 +74,10 @@ print(f"Number of pages: {num_pages}")
 
 Use `mode="static"` if you want to simulate LLM call, without executing LLM backend.
 
+Method `run_inference` will return results and number of pages processed.
+
+Note: GPU backend `katanaml/sparrow-qwen2-vl-7b` is private, to be able to run below command, you need to create your own backend on Hugging Face space using [code](https://github.com/katanaml/sparrow/tree/main/sparrow-data/parse/sparrow_parse/vllm/infra/qwen2_vl_7b) from Sparrow Parse.
+
 ## PDF pre-processing
 
 ```

{sparrow-parse-0.3.6 → sparrow-parse-0.3.8}/sparrow_parse.egg-info/requires.txt

@@ -1,8 +1,11 @@
 rich
-transformers==4.41.2
+transformers==4.45.1
 sentence-transformers==3.0.1
 numpy==1.26.4
 pypdf==4.3.0
 easyocr==1.7.1
 gradio_client
 pdf2image
+
+[:sys_platform == "darwin" and platform_machine == "arm64"]
+mlx-vlm==0.1.1

sparrow-parse-0.3.6/sparrow_parse/__init__.py

@@ -1 +0,0 @@
-__version__ = '0.3.6'

sparrow-parse-0.3.6/sparrow_parse/vllm/huggingface_inference.py

@@ -1,40 +0,0 @@
-from gradio_client import Client, handle_file
-from sparrow_parse.vllm.inference_base import ModelInference
-import json
-
-
-class HuggingFaceInference(ModelInference):
-    def __init__(self, hf_space, hf_token):
-        self.hf_space = hf_space
-        self.hf_token = hf_token
-
-
-    def process_response(self, output_text):
-        json_string = output_text
-
-        json_string = json_string.strip("[]'")
-        json_string = json_string.replace("```json\n", "").replace("\n```", "")
-        json_string = json_string.replace("'", "")
-
-        try:
-            formatted_json = json.loads(json_string)
-            return json.dumps(formatted_json, indent=2)
-        except json.JSONDecodeError as e:
-            print("Failed to parse JSON:", e)
-            return output_text
-
-
-    def inference(self, input_data, mode=None):
-        if mode == "static":
-            simple_json = self.get_simple_json()
-            return simple_json
-
-        client = Client(self.hf_space, hf_token=self.hf_token)
-
-        result = client.predict(
-            image=handle_file(input_data[0]["file_path"]),
-            text_input=input_data[0]["text_input"],
-            api_name="/run_inference"
-        )
-
-        return self.process_response(result)