feat: use thread lock to prevent racing condition (#430)

This PR puts table model initialization behind a threadlock so it is
thread safe.

## testing
Use the following script to test (after changing extension) loading
model in multithreading env.


[test_threading.txt](https://github.com/user-attachments/files/20560179/test_threading.txt)

Author: badGarnet

CHANGELOG.md

@@ -1,5 +1,6 @@
-## 1.0.5-dev0
+## 1.0.5
 
+* feat: add thread lock to prevent racing condition when instantiating singletons
 * feat: parametrize edge config for `DetrImageProcessor` with env variables
 
 ## 1.0.4

test_unstructured_inference/models/test_tables.py

@@ -1,4 +1,6 @@
 import os
+import threading
+from copy import deepcopy
 
 import numpy as np
 import pytest
@@ -7,7 +9,6 @@
 from transformers.models.table_transformer.modeling_table_transformer import (
     TableTransformerDecoder,
 )
-from copy import deepcopy
 
 import unstructured_inference.models.table_postprocess as postprocess
 from unstructured_inference.models import tables
@@ -572,7 +573,7 @@ def test_load_table_model_raises_when_not_available(model_path):
 
 
 @pytest.mark.parametrize(
-    "bbox1, bbox2, expected_result",
+    ("bbox1", "bbox2", "expected_result"),
     [
         ((0, 0, 5, 5), (2, 2, 7, 7), 0.36),
         ((0, 0, 0, 0), (6, 6, 10, 10), 0),
@@ -921,7 +922,9 @@ def test_table_prediction_output_format(
     )
     if output_format:
         result = table_transformer.run_prediction(
-            example_image, result_format=output_format, ocr_tokens=mocked_ocr_tokens
+            example_image,
+            result_format=output_format,
+            ocr_tokens=mocked_ocr_tokens,
         )
     else:
         result = table_transformer.run_prediction(example_image, ocr_tokens=mocked_ocr_tokens)
@@ -952,7 +955,9 @@ def test_table_prediction_output_format_when_wrong_type_then_value_error(
     )
     with pytest.raises(ValueError):
         table_transformer.run_prediction(
-            example_image, result_format="Wrong format", ocr_tokens=mocked_ocr_tokens
+            example_image,
+            result_format="Wrong format",
+            ocr_tokens=mocked_ocr_tokens,
         )
 
 
@@ -991,7 +996,8 @@ def test_table_prediction_with_no_ocr_tokens(table_transformer, example_image):
     ],
 )
 def test_objects_are_filtered_based_on_class_thresholds_when_correct_prediction_and_threshold(
-    thresholds, expected_object_number
+    thresholds,
+    expected_object_number,
 ):
     objects = [
         {"label": "0", "score": 0.2},
@@ -1010,7 +1016,8 @@ def test_objects_are_filtered_based_on_class_thresholds_when_correct_prediction_
     ],
 )
 def test_objects_are_filtered_based_on_class_thresholds_when_two_classes(
-    thresholds, expected_object_number
+    thresholds,
+    expected_object_number,
 ):
     objects = [
         {"label": "0", "score": 0.2},
@@ -1800,7 +1807,7 @@ def test_compute_confidence_score_zero_division_error_handling():
 
 
 @pytest.mark.parametrize(
-    "column_span_score, row_span_score, expected_text_to_indexes",
+    ("column_span_score", "row_span_score", "expected_text_to_indexes"),
     [
         (
             0.9,
@@ -1827,7 +1834,9 @@ def test_compute_confidence_score_zero_division_error_handling():
     ],
 )
 def test_subcells_filtering_when_overlapping_spanning_cells(
-    column_span_score, row_span_score, expected_text_to_indexes
+    column_span_score,
+    row_span_score,
+    expected_text_to_indexes,
 ):
     """
     # table
@@ -1894,3 +1903,17 @@ def test_subcells_filtering_when_overlapping_spanning_cells(
 
     predicted_cells_after_reorder, _ = structure_to_cells(saved_table_structure, tokens=tokens)
     assert predicted_cells_after_reorder == predicted_cells
+
+
+def test_model_init_is_thread_safe():
+    threads = []
+    tables.tables_agent.model = None
+    for i in range(5):
+        thread = threading.Thread(target=tables.load_agent)
+        threads.append(thread)
+        thread.start()
+
+    for thread in threads:
+        thread.join()
+
+    assert tables.tables_agent.model is not None

unstructured_inference/__version__.py

@@ -1 +1 @@
-__version__ = "1.0.5-dev0"  # pragma: no cover
+__version__ = "1.0.5"  # pragma: no cover

unstructured_inference/models/base.py

@@ -2,6 +2,7 @@
 
 import json
 import os
+import threading
 from typing import Dict, Optional, Tuple, Type
 
 from unstructured_inference.models.detectron2onnx import (
@@ -18,12 +19,15 @@
 
 class Models(object):
     _instance = None
+    _lock = threading.Lock()
 
     def __new__(cls):
         """return an instance if one already exists otherwise create an instance"""
         if cls._instance is None:
-            cls._instance = super(Models, cls).__new__(cls)
-            cls.models: Dict[str, UnstructuredModel] = {}
+            with cls._lock:
+                if cls._instance is None:
+                    cls._instance = super(Models, cls).__new__(cls)
+                    cls.models: Dict[str, UnstructuredModel] = {}
         return cls._instance
 
     def __contains__(self, key):

unstructured_inference/models/tables.py

@@ -1,5 +1,6 @@
 # https://github.com/microsoft/table-transformer/blob/main/src/inference.py
 # https://github.com/NielsRogge/Transformers-Tutorials/blob/master/Table%20Transformer/Using_Table_Transformer_for_table_detection_and_table_structure_recognition.ipynb
+import threading
 import xml.etree.ElementTree as ET
 from collections import defaultdict
 from pathlib import Path
@@ -23,20 +24,21 @@
 
 from . import table_postprocess as postprocess
 
+DEFAULT_MODEL = "microsoft/table-transformer-structure-recognition"
+
 
 class UnstructuredTableTransformerModel(UnstructuredModel):
     """Unstructured model wrapper for table-transformer."""
 
     _instance = None
+    _lock = threading.Lock()
 
-    def __init__(self):
-        pass
-
-    @classmethod
-    def instance(cls):
+    def __new__(cls):
         """return an instance if one already exists otherwise create an instance"""
         if cls._instance is None:
-            cls._instance = cls.__new__(cls)
+            with cls._lock:
+                if cls._instance is None:
+                    cls._instance = super(UnstructuredTableTransformerModel, cls).__new__(cls)
         return cls._instance
 
     def predict(
@@ -70,7 +72,7 @@ def initialize(
     ):
         """Loads the donut model using the specified parameters"""
         self.device = device
-        self.feature_extractor = DetrImageProcessor.from_pretrained(model)
+        self.feature_extractor = DetrImageProcessor.from_pretrained(model, device_map=self.device)
         # value not set in the configuration and needed for newer models
         # https://huggingface.co/microsoft/table-transformer-structure-recognition-v1.1-all/discussions/1
         self.feature_extractor.size["shortest_edge"] = inference_config.IMG_PROCESSOR_SHORTEST_EDGE
@@ -145,15 +147,17 @@ def run_prediction(
         return prediction
 
 
-tables_agent: UnstructuredTableTransformerModel = UnstructuredTableTransformerModel.instance()
+tables_agent: UnstructuredTableTransformerModel = UnstructuredTableTransformerModel()
 
 
 def load_agent():
     """Loads the Table agent."""
 
-    if not hasattr(tables_agent, "model"):
-        logger.info("Loading the Table agent ...")
-        tables_agent.initialize("microsoft/table-transformer-structure-recognition")
+    if getattr(tables_agent, "model", None) is None:
+        with tables_agent._lock:
+            if getattr(tables_agent, "model", None) is None:
+                logger.info("Loading the Table agent ...")
+                tables_agent.initialize(DEFAULT_MODEL)
 
     return