v1.5.13

RELEASE.md

@@ -1,5 +1,9 @@
 # Release Notes
 
+## v1.5.13
+
+- ALPRDemo: improve UI experience, fix README typos
+
 ## v1.5.12
 
 - ALPRDemo increase docker compose compatibility

VERSION

@@ -1 +1 @@
-v1.5.12
+v1.5.13

deployment-examples/ALPRDemo/README.md

@@ -40,7 +40,7 @@ Contact [[email protected]](mailto:[email protected]) with any questio
     * Source for it is in `./ui/python`
     * Can (and perhaps should) be ran remotely relatively to the docker-compose.
     * Make sure to install requirements with `pip3 install -r ./ui/python/requirements.txt`
-    * Then run with `cd ./ui/python && python3 ALPRDemo.py`
+    * Then run with `cd ./ui/python && python3 ALPRUI.py`
 
 ## General
 
@@ -59,4 +59,14 @@ If you have an NVIDIA GPU installed and properly configured, you can run the fol
 SIO_DOCKER_RUNTIME=nvidia docker compose up -d
 ```
 
+## OS Compatibility
+
+`SIO_DOCKER_TAG_VARIANT` environment variable used in `docker-compose` controls the flavor of SIO analytics container image. While on x86 systems thing largely work without setting it, on Jetson-based system, set it to the value most compatible with your base OS.
+
+* `-r32.4.3-arm64v8` (built for hardware running Jetpack 4.4)
+* `-r32.7.3-arm64v8` (built for hardware running Jetpack 4.6)
+* `-r35.3.1-arm64v8` (work in progress, built for hardware running Jetpack 5.1)
+* `-amd64` for x86 based systems
+
+This variable may already be pre-set in more recent releases of ShOS / on Sighthound devices.
 

deployment-examples/ALPRDemo/docker-compose.yml

@@ -30,6 +30,23 @@ services:
       - SIO_DATA_DIR=/data/.sio
       # We need this to see output from Python extension module
       - PYTHONUNBUFFERED=1
+      # Way to force specific inference runtime on platforms supporting more than one.
+      # D3V for OpenVino, D3T for TensorRT, D3F for TFLite
+      # - SIO_INFERENCE_RUNTIME=D3V
+      # Detection mode for platforms and engines supporting more than one: 16 for fp16, 8 for int8
+      #   TensorRT and OpenVino only.
+      # - SIO_INFERENCE_MODE=16
+      # Timeout (ms) at which inference will be executed, even if the batch isn't full.
+      # Longer values increase delays, but decrease GPU load.
+      #   TensorRT only.
+      # - SIO_INFERENCE_MAX_INTERBATCH_DELAY=40
+      # Disabling NPP can be useful on systems, where GPU is overloaded (and thus should be preserved for inference),
+      # and a lot of CPU resources are available. Use with care.
+      #   NVidia platforms only.
+      # - SIO_NPP_ENABLE=1
+      # Configured inference batch size. Changing this value will result in re-creating CUDA engine for all the models.
+      #   TensorRT only
+      # - SIO_TENSORRT_RUNTIME_BATCH=16
     # Container runtime defaults to `runc` if SIO_DOCKER_RUNTIME not set. Use `nvidia` if GPU is installed.
     runtime: ${SIO_DOCKER_RUNTIME-runc}
     volumes:

deployment-examples/ALPRDemo/ui/python/ALPRUI.py

@@ -14,6 +14,8 @@
 from datetime import datetime
 from SIOParser import SIOParser
 
+kDefaultTimeout = 5
+
 def epoch_to_offset(epoch_timestamp):
     return datetime.utcfromtimestamp(epoch_timestamp/1000).strftime('%H:%M:%S')
 
@@ -74,6 +76,7 @@ def onSave(self, event):
             self.settings["api_port"] = self.port_text.GetValue()
             self.settings["refresh_rate"] = int(self.refresh_rate_text.GetValue())
             self.settings["max_entries"] = int(self.max_entries_text.GetValue())
+            self.settings["timeout"] = kDefaultTimeout
             self.EndModal(wx.ID_OK)
         except ValueError:
             wx.MessageBox("Please enter valid values for all fields.", "Error", wx.OK | wx.ICON_ERROR)
@@ -90,10 +93,12 @@ def __init__(self, *args, **kw):
             "api_ip": "10.10.10.20",
             "api_port": "8888",
             "refresh_rate": 10,
-            "max_entries": 50
+            "max_entries": 50,
+            "timeout": 10,
         }
 
         self.data = None
+        self.updating = False
 
         self.initUI()
         self.Center()
@@ -212,7 +217,7 @@ def onUpload(self,event):
                     files = {'file': (os.path.basename(filepath), f)}
                     uri = f"{self.apiRoot()}/folderwatch/upload/us"
 
-                    response = requests.post(uri, files=files)
+                    response = requests.post(uri, files=files, timeout=self.settings['timeout'])
 
                     # Check response status
                     if response.status_code != 200:
@@ -228,10 +233,24 @@ def onUpload(self,event):
                     self.uploaded_files_list.InsertItem(idx, filepath)
                     self.uploaded_files_list.SetItem(idx, 1, id)
                     self.uploaded_files_list.SetItem(idx, 2, "Uploaded")
-
+            except requests.exceptions.ConnectionError:
+                self.handleTimeout()
+            except requests.Timeout:
+                self.handleTimeout()
             except Exception as e:
                 wx.MessageBox(f"An error occurred: {str(e)}", "Error", wx.OK | wx.ICON_ERROR)
 
+    # =========================================================================
+    def handleTimeout(self):
+        result = wx.MessageBox(f"Timeout reached communicating to {self.settings['api_ip']}. Do you want to modify settings?", "Update settings?", wx.YES_NO | wx.ICON_QUESTION)
+
+        # Check the response
+        if result == wx.YES:
+            shouldStart = self.auto_refresh
+            self.stopAutoRefresh()
+            self.onSettings(None)
+            if shouldStart:
+                self.startAutoRefresh()
 
     # =========================================================================
     def onTabChanged(self,event):
@@ -401,10 +420,14 @@ def populateRemoteFrame(self, index):
 
         uri = f"{self.apiRoot()}/plates/image/{src}"
         try:
-            response = requests.get(uri,params={'id':id})
+            response = requests.get(uri, params={'id':id}, timeout=self.settings['timeout'])
             if response.status_code != 200:
                 raise Exception(f"Error {response.status_code} retrieving latest results from {uri}")
             self.updateImage(response.content, response.headers['Content-Type'], box)
+        except requests.exceptions.ConnectionError:
+            self.handleTimeout()
+        except requests.Timeout:
+            self.handleTimeout()
         except Exception as e:
             print(f"Error updating image contents: {e}")
             self.clearImage(self.image_ctrl)
@@ -469,6 +492,9 @@ def apiRoot(self):
 
     # =========================================================================
     def updateFileTab(self):
+        if self.updating:
+            return
+        self.updating = True
         item_count = self.uploaded_files_list.GetItemCount()
         for i in range(item_count):
             item_file = self.uploaded_files_list.GetItemText(i)
@@ -484,10 +510,17 @@ def updateFileTab(self):
 
             uri = f"{self.apiRoot()}/folderwatch/status/{item_id}"
 
-            response = requests.get(uri)
+            try:
+                response = requests.get(uri, timeout=self.settings['timeout'])
+            except requests.exceptions.ConnectionError:
+                self.handleTimeout()
+            except requests.Timeout:
+                self.handleTimeout()
+                response = None
+
 
             # Check response status
-            if response.status_code != 200:
+            if not response or response.status_code != 200:
                 self.uploaded_files_list.SetItem(i,2,"error")
                 continue
 
@@ -498,19 +531,29 @@ def updateFileTab(self):
             if status == "completed":
                 result = j.get("result","")
                 self.uploaded_files_results[item_file] = result
+        self.updating = False
+
 
 
     # =========================================================================
     def updateCurrentTab(self):
+        if self.updating:
+            return
+        self.updating = True
         try:
-            response = requests.get(f"{self.apiRoot()}/plates/latest/{self.settings['max_entries']}")
+            response = requests.get(f"{self.apiRoot()}/plates/latest/{self.settings['max_entries']}",
+                                    timeout=self.settings['timeout'])
             if response.status_code != 200:
                 raise Exception(f"Error {response.status_code} retrieving latest results")
-                return
             self.data = response.json()
             self.populateListWithData(self.data, self.list_box, False)
+        except requests.exceptions.ConnectionError:
+            self.handleTimeout()
+        except requests.Timeout:
+            self.handleTimeout()
         except Exception as e:
             print(f"Error updating current tab: {e}")
+        self.updating = False
 
     # =========================================================================
     def onRefreshCurrent(self, event):
@@ -557,11 +600,15 @@ def onSearch(self, event):
         if end:
             uri = uri + f"/{end}"
         try:
-            response = requests.get(uri, params=params)
+            response = requests.get(uri, params=params, timeout=self.settings['timeout'])
             if response.status_code != 200:
                 raise Exception(f"Error {response.status_code} retrieving search results")
             self.data = response.json()
             self.populateListWithData(self.data, self.list_box, False)
+        except requests.exceptions.ConnectionError:
+            self.handleTimeout()
+        except requests.Timeout:
+            self.handleTimeout()
         except Exception as e:
             wx.MessageBox(f"Error updating search tab: {e}", "Error", wx.OK | wx.ICON_ERROR)
 

deployment-examples/SIOOnDemandAnalytics/README.md

@@ -15,4 +15,14 @@ REST API Flask module is bringing it all together, providing a front-end impleme
 ## Client
 
 A sample client is implemented in `SIOOnDemandAnalytics/clients/OnDemandTest.py`.
-The client can be ran using `python3 ./clients/OnDemandTest.py [-i inputImage] [-o outputFolder]`
+The client can be ran using `python3 ./clients/OnDemandTest.py [-i inputImage] [-o outputFolder]`
+
+## OS Compatibility
+
+`SIO_DOCKER_TAG_VARIANT` environment variable used in `docker-compose` controls the flavor of SIO analytics container image. While on x86 systems thing largely work without setting it, on Jetson-based system, set it to the value most compatible with your base OS.
+
+* `-r32.4.3-arm64v8` (built for hardware running Jetpack 4.4)
+* `-r32.7.3-arm64v8` (built for hardware running Jetpack 4.6)
+* `-r35.3.1-arm64v8` (work in progress, built for hardware running Jetpack 5.1)
+* `-amd64` for x86 based systems
+

deployment-examples/SighthoundRestApiGateway/README.md

@@ -34,3 +34,14 @@ We recommend using the UI Demo for development purposes only, and it should be r
 ## Request and Response
 
 See the [online documentation](https://dev.sighthound.com/cloud-api/docs/quickstart/docker/#request-format) for details on the request/response formats and code samples.
+
+## OS Compatibility
+
+`SIO_DOCKER_TAG_VARIANT` environment variable used in `docker-compose` controls the flavor of SIO analytics container image. While on x86 systems thing largely work without setting it, on Jetson-based system, set it to the value most compatible with your base OS.
+
+* `-r32.4.3-arm64v8` (built for hardware running Jetpack 4.4)
+* `-r32.7.3-arm64v8` (built for hardware running Jetpack 4.6)
+* `-r35.3.1-arm64v8` (work in progress, built for hardware running Jetpack 5.1)
+* `-amd64` for x86 based systems
+
+This variable may already be pre-set in more recent releases of ShOS / on Sighthound devices.

deployment-examples/StandaloneSIOWithExtension/README.md

@@ -43,3 +43,13 @@ camera deployed in such non-deterministic way - thus assuming that monitoring th
 ensures we're not going to analyze objects outside of the bottom right quadrant ROI (`lp_roiFilter` accomplishes that), and won't analyze objects below size threshold (`vehicle_SizeFilter`
 and `lp_SizeFilter` filters accomplish that).
 
+
+## OS Compatibility
+
+`SIO_DOCKER_TAG_VARIANT` environment variable used in `docker-compose` controls the flavor of SIO analytics container image. While on x86 systems thing largely work without setting it, on Jetson-based system, set it to the value most compatible with your base OS.
+
+* `-r32.4.3-arm64v8` (built for hardware running Jetpack 4.4)
+* `-r32.7.3-arm64v8` (built for hardware running Jetpack 4.6)
+* `-r35.3.1-arm64v8` (work in progress, built for hardware running Jetpack 5.1)
+* `-amd64` for x86 based systems
+

deployment-examples/VideoStreamsConsumer/README.md

@@ -40,4 +40,15 @@ Once the services are up and running, start the client sample with a `docker com
 
 The following client samples are available for this deployment:
 
-* `SIOOutput` - simple AMQP client listening for and consuming SIO output from the message queue
+* `SIOOutput` - simple AMQP client listening for and consuming SIO output from the message queue
+
+## OS Compatibility
+
+`SIO_DOCKER_TAG_VARIANT` environment variable used in `docker-compose` controls the flavor of SIO analytics container image. While on x86 systems thing largely work without setting it, on Jetson-based system, set it to the value most compatible with your base OS.
+
+* `-r32.4.3-arm64v8` (built for hardware running Jetpack 4.4)
+* `-r32.7.3-arm64v8` (built for hardware running Jetpack 4.6)
+* `-r35.3.1-arm64v8` (work in progress, built for hardware running Jetpack 5.1)
+* `-amd64` for x86 based systems
+
+This variable may already be pre-set in more recent releases of ShOS / on Sighthound devices.

deployment-examples/VideoStreamsRecorder/README.md

@@ -39,3 +39,15 @@ SIO_DOCKER_RUNTIME=nvidia SIO_DOCKER_TAG_VARIANT="-r32.7.3-arm64v8" docker-compo
 
 Once the services are up and running, start the client sample with a `docker compose up` command while inside the relevant sample's folder in `./clients/python/`
 
+
+## OS Compatibility
+
+`SIO_DOCKER_TAG_VARIANT` environment variable used in `docker-compose` controls the flavor of SIO analytics container image. While on x86 systems thing largely work without setting it, on Jetson-based system, set it to the value most compatible with your base OS.
+
+* `-r32.4.3-arm64v8` (built for hardware running Jetpack 4.4)
+* `-r32.7.3-arm64v8` (built for hardware running Jetpack 4.6)
+* `-r35.3.1-arm64v8` (work in progress, built for hardware running Jetpack 5.1)
+* `-amd64` for x86 based systems
+
+This variable may already be pre-set in more recent releases of ShOS / on Sighthound devices.
+

docs/schemas/anypipe/anypipe.html

@@ -5515,7 +5515,7 @@ <h2 class="mb-0">
     <a href="https://github.com/coveooss/json-schema-for-humans">
      json-schema-for-humans
     </a>
-    on 2024-09-17 at 18:23:29 +0000
+    on 2024-11-08 at 23:24:49 +0000
    </p>
   </footer>
  </body>