sparrow-parse 0.4.1__py3-none-any.whl → 0.4.3__py3-none-any.whl

sparrow_parse/__init__.py

@@ -1 +1 @@
-__version__ = '0.4.1'
+__version__ = '0.4.3'

sparrow_parse/extractors/vllm_extractor.py

@@ -54,7 +54,7 @@ class VLLMExtractor(object):
         """
         file_path = input_data[0]["file_path"]
         if tables_only:
-            return [self._extract_tables(model_inference_instance, file_path, input_data, debug, debug_dir)], 1
+            return self._extract_tables(model_inference_instance, file_path, input_data, debug, debug_dir), 1
         else:
             input_data[0]["file_path"] = [file_path]
             results = model_inference_instance.inference(input_data)
@@ -85,7 +85,8 @@ class VLLMExtractor(object):
                 tables_result = self._extract_tables(
                     model_inference_instance, file_path, input_data, debug, debug_dir, page_index=i
                 )
-                results_array.append(tables_result)
+                # Since _extract_tables returns a list with one JSON string, unpack it
+                results_array.extend(tables_result)  # Unpack the single JSON string
         else:
             if debug:
                 print(f"Processing {len(output_files)} pages for inference at once.")
@@ -118,7 +119,15 @@ class VLLMExtractor(object):
             results_array.append(result)
 
         shutil.rmtree(temp_dir, ignore_errors=True)
-        return json.dumps(results_array, indent=4)
+
+        # Merge results_array elements into a single JSON structure
+        merged_results = {"page_tables": results_array}
+
+        # Format the merged results as a JSON string with indentation
+        formatted_results = json.dumps(merged_results, indent=4)
+
+        # Return the formatted JSON string wrapped in a list
+        return [formatted_results]
 
 
     @staticmethod
@@ -166,7 +175,7 @@ if __name__ == "__main__":
     # ]
     #
     # # Now you can run inference without knowing which implementation is used
-    # results_array, num_pages = extractor.run_inference(model_inference_instance, input_data, tables_only=False,
+    # results_array, num_pages = extractor.run_inference(model_inference_instance, input_data, tables_only=True,
     #                                                    generic_query=False,
     #                                                    debug_dir="/Users/andrejb/Work/katana-git/sparrow/sparrow-ml/llm/data/",
     #                                                    debug=True,

sparrow_parse/vllm/mlx_inference.py

@@ -14,14 +14,12 @@ class MLXInference(ModelInference):
 
     def __init__(self, model_name):
         """
-        Initialize the inference class with the given model name and load the model once.
+        Initialize the inference class with the given model name.
 
         :param model_name: Name of the model to load.
         """
-        self.model, self.processor = self._load_model_and_processor(model_name)
-        self.config = self.model.config
-
-        print(f"Loaded model: {model_name}")
+        self.model_name = model_name
+        print(f"MLXInference initialized with model: {model_name}")
 
 
     @staticmethod
@@ -89,6 +87,10 @@ class MLXInference(ModelInference):
         if mode == "static":
             return [self.get_simple_json()]
 
+        # Load the model and processor
+        model, processor = self._load_model_and_processor(self.model_name)
+        config = model.config
+
         # Prepare absolute file paths
         file_paths = self._extract_file_paths(input_data)
 
@@ -103,10 +105,10 @@ class MLXInference(ModelInference):
             ]
 
             # Generate and process response
-            prompt = apply_chat_template(self.processor, self.config, messages)  # Assuming defined
+            prompt = apply_chat_template(processor, config, messages)  # Assuming defined
             response = generate(
-                self.model,
-                self.processor,
+                model,
+                processor,
                 image,
                 prompt,
                 resize_shape=(width, height),

{sparrow_parse-0.4.1.dist-info → sparrow_parse-0.4.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sparrow-parse
-Version: 0.4.1
+Version: 0.4.3
 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

{sparrow_parse-0.4.1.dist-info → sparrow_parse-0.4.3.dist-info}/RECORD

@@ -1,7 +1,7 @@
-sparrow_parse/__init__.py,sha256=8yPI9dbwQUYqhMtA3RfAi5yJOhZBnz-g8966ssrYXiU,21
+sparrow_parse/__init__.py,sha256=udnlByVnFcZDwWir50pEbTU0bIwgBrpNtAiVExFEzu0,21
 sparrow_parse/__main__.py,sha256=Xs1bpJV0n08KWOoQE34FBYn6EBXZA9HIYJKrE4ZdG78,153
 sparrow_parse/extractors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-sparrow_parse/extractors/vllm_extractor.py,sha256=QIg7AMCfw81YHQN6CutF2ipV_DZ3txSGduPIcvQRmiA,7439
+sparrow_parse/extractors/vllm_extractor.py,sha256=ybWpRpDH0YHoYpHkjIJtm7DQoHJBKNsirK2YIAlMvGo,7863
 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
@@ -11,9 +11,9 @@ sparrow_parse/vllm/huggingface_inference.py,sha256=EJnG6PesGKMc_0qGPN8ufE6pSnhAg
 sparrow_parse/vllm/inference_base.py,sha256=4mwGoAY63MB4cHZpV0czTkJWEzimmiTzqqzKmLNzgjw,820
 sparrow_parse/vllm/inference_factory.py,sha256=FTM65O-dW2WZchHOrNN7_Q3-FlVoAc65iSptuuUuClM,1166
 sparrow_parse/vllm/local_gpu_inference.py,sha256=aHoJTejb5xrXjWDIGu5RBQWEyRCOBCB04sMvO2Wyvg8,628
-sparrow_parse/vllm/mlx_inference.py,sha256=xR40qwjIR0HvrN8x58oOq6F4r1hEANRB-9kcokUQHHU,4748
-sparrow_parse-0.4.1.dist-info/METADATA,sha256=4rmJ1CURKtyTs-ZH1eyHn_VptHosJZwhQFB5Fssr5e0,6432
-sparrow_parse-0.4.1.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
-sparrow_parse-0.4.1.dist-info/entry_points.txt,sha256=8CrvTVTTcz1YuZ8aRCYNOH15ZOAaYLlcbYX3t28HwJY,54
-sparrow_parse-0.4.1.dist-info/top_level.txt,sha256=n6b-WtT91zKLyCPZTP7wvne8v_yvIahcsz-4sX8I0rY,14
-sparrow_parse-0.4.1.dist-info/RECORD,,
+sparrow_parse/vllm/mlx_inference.py,sha256=cx-PLXf1t8ro50YALddj70FiR7s0gk_Ddp-I9XlPQQU,4788
+sparrow_parse-0.4.3.dist-info/METADATA,sha256=W7zeOHa09rgn-58aIdTkNOSqBLgpziDF7sZ_059jaoo,6432
+sparrow_parse-0.4.3.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
+sparrow_parse-0.4.3.dist-info/entry_points.txt,sha256=8CrvTVTTcz1YuZ8aRCYNOH15ZOAaYLlcbYX3t28HwJY,54
+sparrow_parse-0.4.3.dist-info/top_level.txt,sha256=n6b-WtT91zKLyCPZTP7wvne8v_yvIahcsz-4sX8I0rY,14
+sparrow_parse-0.4.3.dist-info/RECORD,,