isort

vision_agent/tools/__init__.py

@@ -16,6 +16,8 @@
  clip,
  closest_box_distance,
  closest_mask_distance,
+ countgd_counting,
+ countgd_example_based_counting,
  depth_anything_v2,
  detr_segmentation,
  dpt_hybrid_midas,
@@ -37,13 +39,11 @@
  load_image,
  loca_visual_prompt_counting,
  loca_zero_shot_counting,
- countgd_counting,
- countgd_example_based_counting,
  ocr,
  overlay_bounding_boxes,
+ overlay_counting_results,
  overlay_heat_map,
  overlay_segmentation_masks,
- overlay_counting_results,
  owl_v2,
  save_image,
  save_json,

vision_agent/tools/tool_utils.py

@@ -1,6 +1,6 @@
-import os
 import inspect
 import logging
+import os
 from typing import Any, Callable, Dict, List, MutableMapping, Optional, Tuple
 
 import pandas as pd
@@ -10,10 +10,10 @@
 from requests.adapters import HTTPAdapter
 from urllib3.util.retry import Retry
 
+from vision_agent.tools.tools_types import BoundingBoxes
 from vision_agent.utils.exceptions import RemoteToolCallFailed
 from vision_agent.utils.execute import Error, MimeType
 from vision_agent.utils.type_defs import LandingaiAPIKey
-from vision_agent.tools.tools_types import BoundingBoxes
 
 _LOGGER = logging.getLogger(__name__)
 _LND_API_KEY = os.environ.get("LANDINGAI_API_KEY", LandingaiAPIKey().api_key)

vision_agent/tools/tools.py

@@ -13,26 +13,26 @@
 import numpy as np
 import requests
 from moviepy.editor import ImageSequenceClip
-from PIL import Image, ImageDraw, ImageFont, ImageEnhance
+from PIL import Image, ImageDraw, ImageEnhance, ImageFont
 from pillow_heif import register_heif_opener # type: ignore
 from pytube import YouTube # type: ignore
 
 from vision_agent.clients.landing_public_api import LandingPublicAPI
 from vision_agent.tools.tool_utils import (
+ filter_bboxes_by_threshold,
  get_tool_descriptions,
  get_tool_documentation,
  get_tools_df,
  get_tools_info,
  send_inference_request,
  send_task_inference_request,
- filter_bboxes_by_threshold,
 )
 from vision_agent.tools.tools_types import (
  FineTuning,
  Florence2FtRequest,
  JobStatus,
- PromptTask,
  ODResponseData,
+ PromptTask,
 )
 from vision_agent.utils import extract_frames_from_video
 from vision_agent.utils.exceptions import FineTuneModelIsNotReady

vision_agent/tools/tools_types.py

@@ -1,8 +1,8 @@
 from enum import Enum
+from typing import List, Optional, Tuple, Union
 from uuid import UUID
-from typing import List, Tuple, Optional, Union
 
-from pydantic import BaseModel, ConfigDict, Field, field_serializer, SerializationInfo
+from pydantic import BaseModel, ConfigDict, Field, SerializationInfo, field_serializer
 
 
 class BboxInput(BaseModel):