vision-agent 0.2.70__py3-none-any.whl → 0.2.72__py3-none-any.whl

vision_agent/tools/tool_utils.py

@@ -3,6 +3,9 @@ import os
 from typing import Any, Dict
 
 import requests
+from requests import Session
+from requests.adapters import HTTPAdapter
+from urllib3.util.retry import Retry
 
 from vision_agent.utils.type_defs import LandingaiAPIKey
 
@@ -11,20 +14,50 @@ _LND_API_KEY = LandingaiAPIKey().api_key
 _LND_API_URL = "https://api.staging.landing.ai/v1/agent"
 
 
-def _send_inference_request(
+def send_inference_request(
     payload: Dict[str, Any], endpoint_name: str
 ) -> Dict[str, Any]:
     if runtime_tag := os.environ.get("RUNTIME_TAG", ""):
         payload["runtime_tag"] = runtime_tag
-    res = requests.post(
-        f"{_LND_API_URL}/model/{endpoint_name}",
+    url = f"{_LND_API_URL}/model/{endpoint_name}"
+    session = _create_requests_session(
+        url=url,
+        num_retry=3,
         headers={
             "Content-Type": "application/json",
             "apikey": _LND_API_KEY,
         },
+    )
+    res = requests.post(
+        f"{_LND_API_URL}/model/{endpoint_name}",
         json=payload,
     )
+    res = session.post(url, json=payload)
     if res.status_code != 200:
-        _LOGGER.error(f"Request failed: {res.text}")
-        raise ValueError(f"Request failed: {res.text}")
+        _LOGGER.error(f"Request failed: {res.status_code} {res.text}")
+        raise ValueError(f"Request failed: {res.status_code} {res.text}")
     return res.json()["data"]  # type: ignore
+
+
+def _create_requests_session(
+    url: str, num_retry: int, headers: Dict[str, str]
+) -> Session:
+    """Create a requests session with retry"""
+    session = Session()
+    retries = Retry(
+        total=num_retry,
+        backoff_factor=2,
+        raise_on_redirect=True,
+        raise_on_status=False,
+        allowed_methods=["GET", "POST", "PUT"],
+        status_forcelist=[
+            408,  # Request Timeout
+            429,  # Too Many Requests (ie. rate limiter).
+            502,  # Bad Gateway
+            503,  # Service Unavailable (include cloud circuit breaker)
+            504,  # Gateway Timeout
+        ],
+    )
+    session.mount(url, HTTPAdapter(max_retries=retries if num_retry > 0 else 0))
+    session.headers.update(headers)
+    return session

vision_agent/tools/tools.py

@@ -15,7 +15,7 @@ from moviepy.editor import ImageSequenceClip
 from PIL import Image, ImageDraw, ImageFont
 from pillow_heif import register_heif_opener  # type: ignore
 
-from vision_agent.tools.tool_utils import _send_inference_request
+from vision_agent.tools.tool_utils import send_inference_request
 from vision_agent.utils import extract_frames_from_video
 from vision_agent.utils.execute import FileSerializer, MimeType
 from vision_agent.utils.image_utils import (
@@ -105,7 +105,7 @@ def grounding_dino(
         ),
         "kwargs": {"box_threshold": box_threshold, "iou_threshold": iou_threshold},
     }
-    data: Dict[str, Any] = _send_inference_request(request_data, "tools")
+    data: Dict[str, Any] = send_inference_request(request_data, "tools")
     return_data = []
     for i in range(len(data["bboxes"])):
         return_data.append(
@@ -161,7 +161,7 @@ def owl_v2(
         "tool": "open_vocab_detection",
         "kwargs": {"box_threshold": box_threshold, "iou_threshold": iou_threshold},
     }
-    data: Dict[str, Any] = _send_inference_request(request_data, "tools")
+    data: Dict[str, Any] = send_inference_request(request_data, "tools")
     return_data = []
     for i in range(len(data["bboxes"])):
         return_data.append(
@@ -225,7 +225,7 @@ def grounding_sam(
         "tool": "visual_grounding_segment",
         "kwargs": {"box_threshold": box_threshold, "iou_threshold": iou_threshold},
     }
-    data: Dict[str, Any] = _send_inference_request(request_data, "tools")
+    data: Dict[str, Any] = send_inference_request(request_data, "tools")
     return_data = []
     for i in range(len(data["bboxes"])):
         return_data.append(
@@ -341,7 +341,7 @@ def loca_zero_shot_counting(image: np.ndarray) -> Dict[str, Any]:
         "image": image_b64,
         "tool": "zero_shot_counting",
     }
-    resp_data = _send_inference_request(data, "tools")
+    resp_data = send_inference_request(data, "tools")
     resp_data["heat_map"] = np.array(b64_to_pil(resp_data["heat_map"][0]))
     return resp_data
 
@@ -376,7 +376,7 @@ def loca_visual_prompt_counting(
         "prompt": bbox_str,
         "tool": "few_shot_counting",
     }
-    resp_data = _send_inference_request(data, "tools")
+    resp_data = send_inference_request(data, "tools")
     resp_data["heat_map"] = np.array(b64_to_pil(resp_data["heat_map"][0]))
     return resp_data
 
@@ -407,7 +407,7 @@ def git_vqa_v2(prompt: str, image: np.ndarray) -> str:
         "tool": "image_question_answering",
     }
 
-    answer = _send_inference_request(data, "tools")
+    answer = send_inference_request(data, "tools")
     return answer["text"][0]  # type: ignore
 
 
@@ -436,7 +436,7 @@ def clip(image: np.ndarray, classes: List[str]) -> Dict[str, Any]:
         "image": image_b64,
         "tool": "closed_set_image_classification",
     }
-    resp_data = _send_inference_request(data, "tools")
+    resp_data = send_inference_request(data, "tools")
     resp_data["scores"] = [round(prob, 4) for prob in resp_data["scores"]]
     return resp_data
 
@@ -463,7 +463,7 @@ def vit_image_classification(image: np.ndarray) -> Dict[str, Any]:
         "image": image_b64,
         "tool": "image_classification",
     }
-    resp_data = _send_inference_request(data, "tools")
+    resp_data = send_inference_request(data, "tools")
     resp_data["scores"] = [round(prob, 4) for prob in resp_data["scores"]]
     return resp_data
 
@@ -490,7 +490,7 @@ def vit_nsfw_classification(image: np.ndarray) -> Dict[str, Any]:
         "image": image_b64,
         "tool": "nsfw_image_classification",
     }
-    resp_data = _send_inference_request(data, "tools")
+    resp_data = send_inference_request(data, "tools")
     resp_data["scores"] = round(resp_data["scores"], 4)
     return resp_data
 
@@ -517,7 +517,7 @@ def blip_image_caption(image: np.ndarray) -> str:
         "tool": "image_captioning",
     }
 
-    answer = _send_inference_request(data, "tools")
+    answer = send_inference_request(data, "tools")
     return answer["text"][0]  # type: ignore
 
 

{vision_agent-0.2.70.dist-info → vision_agent-0.2.72.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vision-agent
-Version: 0.2.70
+Version: 0.2.72
 Summary: Toolset for Vision Agent
 Author: Landing AI
 Author-email: dev@landing.ai
@@ -78,7 +78,7 @@ export OPENAI_API_KEY="your-api-key"
 ```
 
 ### Important Note on API Usage
-Please be aware that using the API in this project requires you to have API credits (minimum of five US dollars). This is different from the OpenAI subscription used in this chatbot. If you don't have credit, further information can be found [here](https://github.com/landing-ai/vision-agent?tab=readme-ov-file#to-get-started-with-api-credits)
+Please be aware that using the API in this project requires you to have API credits (minimum of five US dollars). This is different from the OpenAI subscription used in this chatbot. If you don't have credit, further information can be found [here](https://github.com/landing-ai/vision-agent?tab=readme-ov-file#how-to-get-started-with-openai-api-credits)
 
 ### Vision Agent
 #### Basic Usage

{vision_agent-0.2.70.dist-info → vision_agent-0.2.72.dist-info}/RECORD

@@ -9,15 +9,15 @@ vision_agent/lmm/__init__.py,sha256=bw24xyQJHGzmph5e-bKCiTh9AX6tRFI2OUd0mofxjZI,
 vision_agent/lmm/lmm.py,sha256=V7jfU94HwA-SiQLY14USHrSGtagVKCNGjZhW5MyKipo,11547
 vision_agent/tools/__init__.py,sha256=aE1O8cMeLDPO50Sc-CuAQ_Akh0viz7vBxDcVeZNqsA0,1604
 vision_agent/tools/prompts.py,sha256=V1z4YJLXZuUl_iZ5rY0M5hHc_2tmMEUKr0WocXKGt4E,1430
-vision_agent/tools/tool_utils.py,sha256=wzRacbUpqk9hhfX_Y08rL8qP0XCN2w-8IZoYLi3Upn4,869
-vision_agent/tools/tools.py,sha256=Qzwm_wu6KJh-3DSoNmZ4Lv8jCCNJMwKIPBFxxN6FmDo,31397
+vision_agent/tools/tool_utils.py,sha256=ELcCD05xmER9KRN1lt5VbI31Kyvp-8G-ERSwSleT3Ic,1956
+vision_agent/tools/tools.py,sha256=TaDZIvYsYNleqDsETfoZiPWPBZjyimXhudLdFZ5NsLE,31386
 vision_agent/utils/__init__.py,sha256=CW84HnhqI6XQVuxf2KifkLnSuO7EOhmuL09-gAymAak,219
 vision_agent/utils/execute.py,sha256=QImS69SN00logF-E68aNpT7YsJVRQOhZYlNLmCNEfro,21337
 vision_agent/utils/image_utils.py,sha256=_cdiS5YrLzqkq_ZgFUO897m5M4_SCIThwUy4lOklfB8,7700
 vision_agent/utils/sim.py,sha256=ci6Eta73dDgLP1Ajtknbgmf1g8aAvBHqlVQvBuLMKXQ,4427
 vision_agent/utils/type_defs.py,sha256=BlI8ywWHAplC7kYWLvt4AOdnKpEW3qWEFm-GEOSkrFQ,1792
 vision_agent/utils/video.py,sha256=rNmU9KEIkZB5-EztZNlUiKYN0mm_55A_2VGUM0QpqLA,8779
-vision_agent-0.2.70.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-vision_agent-0.2.70.dist-info/METADATA,sha256=rMFWipJt_VyOfG2x7-zSZ-vQXZsZrerGlSrKwy_YlE0,9394
-vision_agent-0.2.70.dist-info/WHEEL,sha256=7Z8_27uaHI_UZAc4Uox4PpBhQ9Y5_modZXWMxtUi4NU,88
-vision_agent-0.2.70.dist-info/RECORD,,
+vision_agent-0.2.72.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+vision_agent-0.2.72.dist-info/METADATA,sha256=OsGet3c8LbdAW475Ln3srl4NINT929DhkFEF6PSSjVI,9405
+vision_agent-0.2.72.dist-info/WHEEL,sha256=7Z8_27uaHI_UZAc4Uox4PpBhQ9Y5_modZXWMxtUi4NU,88
+vision_agent-0.2.72.dist-info/RECORD,,