Merge branch 'main' into cf-e2e

Author: aseembits93

README.md

@@ -54,7 +54,7 @@ pip install inference-cli && inference server start --dev
 
 This will pull the proper image for your machine and start it in development mode.
 
-In development mode, a Jupyter notebook server with a quickstart guide runs on 
+In development mode, a Jupyter notebook server with a quickstart guide runs on
 [http://localhost:9001/notebook/start](http://localhost:9001/notebook/start). Dive in there for a whirlwind tour
 of your new Inference Server's functionality!
 
@@ -86,7 +86,7 @@ Workflows allow you to extend simple model predictions to build computer vision
     <!-- Left cell (thumbnail) -->
     <td width="300" valign="top">
       <a href="https://youtu.be/aPxlImNxj5A">
-        <img src="https://img.youtube.com/vi/aPxlImNxj5A/0.jpg" 
+        <img src="https://img.youtube.com/vi/aPxlImNxj5A/0.jpg"
              alt="Self Checkout with Workflows" width="300" />
       </a>
     </td>
@@ -104,7 +104,7 @@ Workflows allow you to extend simple model predictions to build computer vision
   <tr>
     <td width="300" valign="top">
       <a href="https://youtu.be/r3Ke7ZEh2Qo">
-        <img src="https://img.youtube.com/vi/r3Ke7ZEh2Qo/0.jpg" 
+        <img src="https://img.youtube.com/vi/r3Ke7ZEh2Qo/0.jpg"
              alt="Workflows Tutorial" width="300" />
       </a>
     </td>
@@ -116,7 +116,7 @@ Workflows allow you to extend simple model predictions to build computer vision
       </strong><br />
       <strong>Created: 6 Jan 2025</strong><br /><br />
       Learn how to build and deploy Workflows for common use-cases like detecting
-      vehicles, filtering detections, visualizing results, and calculating dwell 
+      vehicles, filtering detections, visualizing results, and calculating dwell
       time on a live video stream.
     </td>
   </tr>
@@ -125,7 +125,7 @@ Workflows allow you to extend simple model predictions to build computer vision
     <!-- Left cell (thumbnail) -->
     <td width="300" valign="top">
       <a href="https://youtu.be/tZa-QgFn7jg">
-        <img src="https://img.youtube.com/vi/tZa-QgFn7jg/0.jpg" 
+        <img src="https://img.youtube.com/vi/tZa-QgFn7jg/0.jpg"
              alt="Smart Parking with AI" width="300" />
       </a>
     </td>
@@ -143,7 +143,7 @@ Workflows allow you to extend simple model predictions to build computer vision
 </table>
 
 ## 📟 connecting via api
-  
+
 Once you've installed Inference, your machine is a fully-featured CV center.
 You can use its API to run models and workflows on images and video streams.
 By default, the server is running locally on
@@ -334,42 +334,36 @@ We would love your input to improve Roboflow Inference! Please see our [contribu
           <img
             src="https://media.roboflow.com/notebooks/template/icons/purple/youtube.png?ik-sdk-version=javascript-1.4.3&updatedAt=1672949634652"
             width="3%"
-          />
-      </a>
+          /></a>
       <img src="https://raw.githubusercontent.com/ultralytics/assets/main/social/logo-transparent.png" width="3%"/>
       <a href="https://roboflow.com">
           <img
             src="https://media.roboflow.com/notebooks/template/icons/purple/roboflow-app.png?ik-sdk-version=javascript-1.4.3&updatedAt=1672949746649"
             width="3%"
-          />
-      </a>
+          /></a>
       <img src="https://raw.githubusercontent.com/ultralytics/assets/main/social/logo-transparent.png" width="3%"/>
       <a href="https://www.linkedin.com/company/roboflow-ai/">
           <img
             src="https://media.roboflow.com/notebooks/template/icons/purple/linkedin.png?ik-sdk-version=javascript-1.4.3&updatedAt=1672949633691"
             width="3%"
-          />
-      </a>
+          /></a>
       <img src="https://raw.githubusercontent.com/ultralytics/assets/main/social/logo-transparent.png" width="3%"/>
       <a href="https://docs.roboflow.com">
           <img
             src="https://media.roboflow.com/notebooks/template/icons/purple/knowledge.png?ik-sdk-version=javascript-1.4.3&updatedAt=1672949634511"
             width="3%"
-          />
-      </a>
+          /></a>
       <img src="https://raw.githubusercontent.com/ultralytics/assets/main/social/logo-transparent.png" width="3%"/>
       <a href="https://disuss.roboflow.com">
           <img
             src="https://media.roboflow.com/notebooks/template/icons/purple/forum.png?ik-sdk-version=javascript-1.4.3&updatedAt=1672949633584"
             width="3%"
-          />
+          /></a>
       <img src="https://raw.githubusercontent.com/ultralytics/assets/main/social/logo-transparent.png" width="3%"/>
       <a href="https://blog.roboflow.com">
           <img
             src="https://media.roboflow.com/notebooks/template/icons/purple/blog.png?ik-sdk-version=javascript-1.4.3&updatedAt=1672949633605"
             width="3%"
-          />
-      </a>
-      </a>
+          /></a>
   </div>
 </div>

docs/install/index.md

@@ -16,7 +16,7 @@ Simply download the latest installer for your operating system.  You can find th
  - [Download the Roboflow Inference DMG](https://github.com/roboflow/inference/releases) disk image
  - Mount hte disk image by double clicking it
  - Drag the Roboflow Inference App to the Application Folder
- - Go to your Application Folder and double click the Roboflow Infernce App to start the server
+ - Go to your Application Folder and double click the Roboflow Inference App to start the server
 
 
 
@@ -34,7 +34,7 @@ notifications).
 
 ## Run via Docker
 
-The preferred way to use Inference is via Docker 
+The preferred way to use Inference is via Docker
 (see [Why Docker](/understand/architecture.md#why-docker)).
 
 [Install Docker](https://docs.docker.com/engine/install/) (and
@@ -46,7 +46,7 @@ pip install inference-cli
 inference server start
 ```
 
-The `inference server start` command attempts to automatically choose 
+The `inference server start` command attempts to automatically choose
 and configure the optimal container to optimize performance on your machine.
 See [Using Your New Server](#using-your-new-server) for next steps.
 
@@ -57,7 +57,7 @@ See [Using Your New Server](#using-your-new-server) for next steps.
 ## Dev Mode
 
 The `--dev` parameter to `inference server start` starts in development mode.
-This spins up a companion Jupyter notebook server with a quickstart guide on 
+This spins up a companion Jupyter notebook server with a quickstart guide on
 [`localhost:9002`](http://localhost:9002). Dive in there for a whirlwind tour
 of your new Inference Server's functionality!
 

docs/install/mac.md

@@ -2,16 +2,16 @@
 
 ## OSX Native App (Apple Silicon)
 
-You can now run Roboflow Inference Server on your Apple Silicon Mac using our native desktop app! 
+You can now run Roboflow Inference Server on your Apple Silicon Mac using our native desktop app!
 
-Simply download the latest DMS disk image from the latest release on Github.  
+Simply download the latest DMS disk image from the latest release on Github.
 ➡️ **[View Latest Release and Download Installers on Github](https://github.com/roboflow/inference/releases)**
 
-### OSX Installation Steps 
+### OSX Installation Steps
  - [Download the Roboflow Inference DMG](https://github.com/roboflow/inference/releases) disk image
  - Mount hte disk image by double clicking it
  - Drag the Roboflow Inference App to the Application Folder
- - Go to your Application Folder and double click the Roboflow Infernce App to start the server
+ - Go to your Application Folder and double click the Roboflow Inference App to start the server
 
 ## Using Docker
 === "CPU"

inference/core/interfaces/stream_manager/manager_app/communication.py

@@ -29,15 +29,15 @@ def receive_socket_data(
             private_message=f"Header is indicating non positive payload size: {payload_size}",
             public_message=f"Header is indicating non positive payload size: {payload_size}",
         )
-    received = b""
+    received = bytearray()
     while len(received) < payload_size:
         chunk = source.recv(buffer_size)
         if len(chunk) == 0:
             raise TransmissionChannelClosed(
                 private_message="Socket was closed to read before payload was decoded.",
                 public_message="Socket was closed to read before payload was decoded.",
             )
-        received += chunk
+        received.extend(chunk)
     try:
         return json.loads(received)
     except ValueError as error:

inference/core/workflows/core_steps/fusion/detections_consensus/v1.py

@@ -2,6 +2,7 @@
 import statistics
 from collections import Counter
 from enum import Enum
+from functools import lru_cache
 from typing import Dict, Generator, List, Literal, Optional, Set, Tuple, Type, Union
 from uuid import uuid4
 
@@ -176,6 +177,7 @@ def get_parameters_accepting_batches(cls) -> List[str]:
         return ["predictions_batches"]
 
     @classmethod
+    @lru_cache(maxsize=None)
     def describe_outputs(cls) -> List[OutputDefinition]:
         return [
             OutputDefinition(

inference/models/rfdetr/rfdetr.py

@@ -316,10 +316,12 @@ def postprocess(
 
                 boxes_xyxy = boxes_input / scale
 
-            boxes_xyxy[:, 0] = np.clip(boxes_xyxy[:, 0], 0, orig_w)
-            boxes_xyxy[:, 1] = np.clip(boxes_xyxy[:, 1], 0, orig_h)
-            boxes_xyxy[:, 2] = np.clip(boxes_xyxy[:, 2], 0, orig_w)
-            boxes_xyxy[:, 3] = np.clip(boxes_xyxy[:, 3], 0, orig_h)
+            np.clip(
+                boxes_xyxy,
+                [0, 0, 0, 0],
+                [orig_w, orig_h, orig_w, orig_h],
+                out=boxes_xyxy,
+            )
 
             batch_predictions = np.column_stack(
                 (

requirements/requirements.sam.txt

@@ -1,5 +1,5 @@
 rf-segment-anything==1.0
 samv2==0.0.4
 rasterio~=1.4.0
-torch>=2.0.1,<2.7.0
+torch>=2.0.1,<2.8.0
 torchvision>=0.15.2

requirements/requirements.transformers.txt

@@ -1,4 +1,4 @@
-torch>=2.0.1,<2.7.0
+torch>=2.0.1,<2.8.0
 torchvision>=0.15.0
 transformers>=4.50.0,<4.52.0
 timm~=1.0.0