sparrow-parse 0.3.6__py3-none-any.whl → 0.3.7__py3-none-any.whl

sparrow_parse/__init__.py

@@ -1 +1 @@
-__version__ = '0.3.6'
+__version__ = '0.3.7'

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/vllm/huggingface_inference.py

@@ -1,6 +1,8 @@
 from gradio_client import Client, handle_file
 from sparrow_parse.vllm.inference_base import ModelInference
 import json
+import os
+import ast
 
 
 class HuggingFaceInference(ModelInference):
@@ -27,14 +29,32 @@ class HuggingFaceInference(ModelInference):
     def inference(self, input_data, mode=None):
         if mode == "static":
             simple_json = self.get_simple_json()
-            return 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"
+        # 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
         )
 
-        return self.process_response(result)
+        # 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.dist-info → sparrow_parse-0.3.7.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sparrow-parse
-Version: 0.3.6
+Version: 0.3.7
 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
@@ -73,7 +73,7 @@ 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)
@@ -82,6 +82,10 @@ 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.dist-info → sparrow_parse-0.3.7.dist-info}/RECORD

@@ -1,18 +1,18 @@
-sparrow_parse/__init__.py,sha256=IbpUPwvtjLOqowcOFsWQ6LKq-FH6cI19IpvfQlxufq0,21
+sparrow_parse/__init__.py,sha256=V3RDzgFfGW_qKkRklGT6eISHLybQsgfScnd9neXG7Cs,21
 sparrow_parse/__main__.py,sha256=Xs1bpJV0n08KWOoQE34FBYn6EBXZA9HIYJKrE4ZdG78,153
 sparrow_parse/extractors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-sparrow_parse/extractors/vllm_extractor.py,sha256=Wo8sOvsQt6YHd7bvB_DB8MUa71FioO9xcQOWA3PQ6eU,3415
+sparrow_parse/extractors/vllm_extractor.py,sha256=mBPgeyMuHUa6jN_OZLVE-426tD4zYnFT61oxebk7XJc,3191
 sparrow_parse/helpers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sparrow_parse/helpers/pdf_optimizer.py,sha256=GIqQYWtixFeZGCRFXL0lQfQByapCDuQzzRHAkzcPwLE,3302
 sparrow_parse/processors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sparrow_parse/processors/table_structure_processor.py,sha256=bG_6jx66n_KNdY_O6hrZD1D4DHX5Qy__RYcKHmrSGnc,23894
 sparrow_parse/vllm/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-sparrow_parse/vllm/huggingface_inference.py,sha256=nalmPJFfrFlRnfd4yTq4HvIwDvIXjhKUlEyZ6gzMqe0,1239
+sparrow_parse/vllm/huggingface_inference.py,sha256=EJnG6PesGKMc_0qGPN8ufE6pSnhAgFu0XjCbaLCNVyM,1980
 sparrow_parse/vllm/inference_base.py,sha256=4mwGoAY63MB4cHZpV0czTkJWEzimmiTzqqzKmLNzgjw,820
 sparrow_parse/vllm/inference_factory.py,sha256=r04e95uPWG5l8Q23yeDqKmvFxLyF991aA2m0hfBTNn8,993
 sparrow_parse/vllm/local_gpu_inference.py,sha256=aHoJTejb5xrXjWDIGu5RBQWEyRCOBCB04sMvO2Wyvg8,628
-sparrow_parse-0.3.6.dist-info/METADATA,sha256=ANS8eWCx07bQOOFFnJUKwsiPo-ZT42b8DvMwP9o-jf4,5827
-sparrow_parse-0.3.6.dist-info/WHEEL,sha256=eOLhNAGa2EW3wWl_TU484h7q1UNgy0JXjjoqKoxAAQc,92
-sparrow_parse-0.3.6.dist-info/entry_points.txt,sha256=8CrvTVTTcz1YuZ8aRCYNOH15ZOAaYLlcbYX3t28HwJY,54
-sparrow_parse-0.3.6.dist-info/top_level.txt,sha256=n6b-WtT91zKLyCPZTP7wvne8v_yvIahcsz-4sX8I0rY,14
-sparrow_parse-0.3.6.dist-info/RECORD,,
+sparrow_parse-0.3.7.dist-info/METADATA,sha256=ErE4fDTkcyOrVbgpc6x9AO9cU3Gf8HbEGsbKmK-F0RA,6187
+sparrow_parse-0.3.7.dist-info/WHEEL,sha256=eOLhNAGa2EW3wWl_TU484h7q1UNgy0JXjjoqKoxAAQc,92
+sparrow_parse-0.3.7.dist-info/entry_points.txt,sha256=8CrvTVTTcz1YuZ8aRCYNOH15ZOAaYLlcbYX3t28HwJY,54
+sparrow_parse-0.3.7.dist-info/top_level.txt,sha256=n6b-WtT91zKLyCPZTP7wvne8v_yvIahcsz-4sX8I0rY,14
+sparrow_parse-0.3.7.dist-info/RECORD,,