Hi all,

We have a project where we communicate with a device using gRPC. On the computer, we run Python in a virtual enviornment. We load that environment in matlab using pyenv. We use MATLAB 2023b (23.2.0.2599560 (R2023b) Update 8) and Python 3.10.11.
Connecting with the device and querying some device information works in both, Python and MATLAB. However, since 2-3 months, we are unable to access members of protobuf messages. Below is an example of the code we execute in Python and the equivalent code executed in MATLAB:
Python:
% C:workdatakingfisher-py.venvScriptspython.exe "…"
import kingfisher_py.lib as kgfLib
scn = kgfLib.device.Scanner(‘10.10.1.1’, ‘8081’)
device_info = scn.device_info.get_info()
device_info.sw_rev
items {
key: "mcu"
value: "v1.0.1"
}
items {
key: "kingfisher"
value: "5.2.0-cam-cal-third-party-rc.1-5-gd6f7161"
}
% response continues…

MATLAB:
>> kgfLib = py.importlib.import_module(‘kingfisher_py.lib’);
>> scn = kgfLib.device.Scanner(‘10.10.1.1’, ‘8081’);
>> deviceInfo = scn.device_info.get_info();
>> deviceInfo.HasField(‘sw_rev’)

ans =

logical

1

>> deviceInfo.sw_rev
Unrecognized method, property, or field ‘sw_rev’ for class ‘py.kf.api.messages.system_pb2.GetDeviceInfoResponse’.

>> deviceInfo

deviceInfo =

Python GetDeviceInfoResponse with properties:

DESCRIPTOR: [1×1 py.google._upb._message.Descriptor]

manufact_rev {
device_name: "BLK360-2060047"
serial_number: "2060047"
}
sw_rev {
items {
key: "mcu"
value: "v1.0.1"
}
items {
key: "kingfisher"
value: "5.2.0-cam-cal-third-party-rc.1-5-gd6f7161"
}
% response continues…

Note that the same virtual environment is active. MATLAB even says that the field sw_rev exists, but still cannot access it. Also, we checked different versions and combinations of MATLAB and Python. Specifically:
Matlab 9.10.0.2198249 (R2021a) Update 8 + Python 3.8.10
Matlab 9.13.0.2193358 (R2022b) Update 5 + Python 3.8.10
Matlab 9.13.0.2193358 (R2022b) Update 5 + Python 3.10.11
Matlab 23.2.0.2599560 (R2023b) Update 8 + Python 3.10.11
The behaviour is the same with all versions. As the whole communication with the device is set up using protobuf, there is not much we can do with the device at this point as we run into this problem all accorss our MATLAB code base.

Is this a known issue, e.g. with a newer protobuf version? As mentioned, we did not have any issues like this until 2-3 months ago.

Hope to get some help or at least an explanation. Thank you :-)Hi all,

We have a project where we communicate with a device using gRPC. On the computer, we run Python in a virtual enviornment. We load that environment in matlab using pyenv. We use MATLAB 2023b (23.2.0.2599560 (R2023b) Update 8) and Python 3.10.11.
Connecting with the device and querying some device information works in both, Python and MATLAB. However, since 2-3 months, we are unable to access members of protobuf messages. Below is an example of the code we execute in Python and the equivalent code executed in MATLAB:
Python:
% C:workdatakingfisher-py.venvScriptspython.exe "…"
import kingfisher_py.lib as kgfLib
scn = kgfLib.device.Scanner(‘10.10.1.1’, ‘8081’)
device_info = scn.device_info.get_info()
device_info.sw_rev
items {
key: "mcu"
value: "v1.0.1"
}
items {
key: "kingfisher"
value: "5.2.0-cam-cal-third-party-rc.1-5-gd6f7161"
}
% response continues…

MATLAB:
>> kgfLib = py.importlib.import_module(‘kingfisher_py.lib’);
>> scn = kgfLib.device.Scanner(‘10.10.1.1’, ‘8081’);
>> deviceInfo = scn.device_info.get_info();
>> deviceInfo.HasField(‘sw_rev’)

ans =

logical

1

>> deviceInfo.sw_rev
Unrecognized method, property, or field ‘sw_rev’ for class ‘py.kf.api.messages.system_pb2.GetDeviceInfoResponse’.

>> deviceInfo

deviceInfo =

Python GetDeviceInfoResponse with properties:

DESCRIPTOR: [1×1 py.google._upb._message.Descriptor]

manufact_rev {
device_name: "BLK360-2060047"
serial_number: "2060047"
}
sw_rev {
items {
key: "mcu"
value: "v1.0.1"
}
items {
key: "kingfisher"
value: "5.2.0-cam-cal-third-party-rc.1-5-gd6f7161"
}
% response continues…

Note that the same virtual environment is active. MATLAB even says that the field sw_rev exists, but still cannot access it. Also, we checked different versions and combinations of MATLAB and Python. Specifically:
Matlab 9.10.0.2198249 (R2021a) Update 8 + Python 3.8.10
Matlab 9.13.0.2193358 (R2022b) Update 5 + Python 3.8.10
Matlab 9.13.0.2193358 (R2022b) Update 5 + Python 3.10.11
Matlab 23.2.0.2599560 (R2023b) Update 8 + Python 3.10.11
The behaviour is the same with all versions. As the whole communication with the device is set up using protobuf, there is not much we can do with the device at this point as we run into this problem all accorss our MATLAB code base.

Is this a known issue, e.g. with a newer protobuf version? As mentioned, we did not have any issues like this until 2-3 months ago.

Hope to get some help or at least an explanation. Thank you 🙂 Hi all,

We have a project where we communicate with a device using gRPC. On the computer, we run Python in a virtual enviornment. We load that environment in matlab using pyenv. We use MATLAB 2023b (23.2.0.2599560 (R2023b) Update 8) and Python 3.10.11.
Connecting with the device and querying some device information works in both, Python and MATLAB. However, since 2-3 months, we are unable to access members of protobuf messages. Below is an example of the code we execute in Python and the equivalent code executed in MATLAB:
Python:
% C:workdatakingfisher-py.venvScriptspython.exe "…"
import kingfisher_py.lib as kgfLib
scn = kgfLib.device.Scanner(‘10.10.1.1’, ‘8081’)
device_info = scn.device_info.get_info()
device_info.sw_rev
items {
key: "mcu"
value: "v1.0.1"
}
items {
key: "kingfisher"
value: "5.2.0-cam-cal-third-party-rc.1-5-gd6f7161"
}
% response continues…

MATLAB:
>> kgfLib = py.importlib.import_module(‘kingfisher_py.lib’);
>> scn = kgfLib.device.Scanner(‘10.10.1.1’, ‘8081’);
>> deviceInfo = scn.device_info.get_info();
>> deviceInfo.HasField(‘sw_rev’)

ans =

logical

1

>> deviceInfo.sw_rev
Unrecognized method, property, or field ‘sw_rev’ for class ‘py.kf.api.messages.system_pb2.GetDeviceInfoResponse’.

>> deviceInfo

deviceInfo =

Python GetDeviceInfoResponse with properties:

DESCRIPTOR: [1×1 py.google._upb._message.Descriptor]

manufact_rev {
device_name: "BLK360-2060047"
serial_number: "2060047"
}
sw_rev {
items {
key: "mcu"
value: "v1.0.1"
}
items {
key: "kingfisher"
value: "5.2.0-cam-cal-third-party-rc.1-5-gd6f7161"
}
% response continues…

Note that the same virtual environment is active. MATLAB even says that the field sw_rev exists, but still cannot access it. Also, we checked different versions and combinations of MATLAB and Python. Specifically:
Matlab 9.10.0.2198249 (R2021a) Update 8 + Python 3.8.10
Matlab 9.13.0.2193358 (R2022b) Update 5 + Python 3.8.10
Matlab 9.13.0.2193358 (R2022b) Update 5 + Python 3.10.11
Matlab 23.2.0.2599560 (R2023b) Update 8 + Python 3.10.11
The behaviour is the same with all versions. As the whole communication with the device is set up using protobuf, there is not much we can do with the device at this point as we run into this problem all accorss our MATLAB code base.

Is this a known issue, e.g. with a newer protobuf version? As mentioned, we did not have any issues like this until 2-3 months ago.

Hope to get some help or at least an explanation. Thank you 🙂 python, grpc, protobuf MATLAB Answers — New Questions

​

When i use yolov4 detector to achieve a single-class target detection tasks, i can only get training-loss and validation-loss after training process.I have tried to use "options" to get metrics ，but it made Errow, wrong use " nnet.cnn.TrainingOptions/validateDataFormatsAndMetricsAreEmpty ,trainYOLOv4ObjectDetector can’t support ‘Metrics’ training options",my code and errow are below:

Then how to get metrics during yolov4 detector training process?
———————————————————————————————-
Meanwhile, after fine-tuning options parameters,I have trained the detector many times, the final training loss maintains 0.1 while validation loss can only get 2.0, how can I avoid this data overfitting?I’ve tried many methods , if you can give me some suggestions , it will be very helpful .When i use yolov4 detector to achieve a single-class target detection tasks, i can only get training-loss and validation-loss after training process.I have tried to use "options" to get metrics ，but it made Errow, wrong use " nnet.cnn.TrainingOptions/validateDataFormatsAndMetricsAreEmpty ,trainYOLOv4ObjectDetector can’t support ‘Metrics’ training options",my code and errow are below:

Then how to get metrics during yolov4 detector training process?
———————————————————————————————-
Meanwhile, after fine-tuning options parameters,I have trained the detector many times, the final training loss maintains 0.1 while validation loss can only get 2.0, how can I avoid this data overfitting?I’ve tried many methods , if you can give me some suggestions , it will be very helpful . When i use yolov4 detector to achieve a single-class target detection tasks, i can only get training-loss and validation-loss after training process.I have tried to use "options" to get metrics ，but it made Errow, wrong use " nnet.cnn.TrainingOptions/validateDataFormatsAndMetricsAreEmpty ,trainYOLOv4ObjectDetector can’t support ‘Metrics’ training options",my code and errow are below:

Then how to get metrics during yolov4 detector training process?
———————————————————————————————-
Meanwhile, after fine-tuning options parameters,I have trained the detector many times, the final training loss maintains 0.1 while validation loss can only get 2.0, how can I avoid this data overfitting?I’ve tried many methods , if you can give me some suggestions , it will be very helpful . yolov4 detector, metrics, overfitting MATLAB Answers — New Questions

​

% Create Environment
env = MYEnv();
% Define State and Action Specifications
stateSpec = env.getObservationInfo();
actionSpec = env.getActionInfo();
stateName = stateSpec.Name;
% Create Actor Network
actorNetwork = [
featureInputLayer(stateSpec.Dimension(1), ‘Name’, stateName)
fullyConnectedLayer(400, ‘Name’, ‘ActorHiddenLayer1’)
reluLayer(‘Name’, ‘ActorReLU1’)
fullyConnectedLayer(300, ‘Name’, ‘ActorHiddenLayer2’)
reluLayer(‘Name’, ‘ActorReLU2’)
fullyConnectedLayer(actionSpec.Dimension(1), ‘Name’, ‘ActorOutputLayer’)
tanhLayer(‘Name’, ‘ActorTanh’)
];
% Create Critic Network
criticNetwork = [
featureInputLayer(stateSpec.Dimension(1), ‘Name’, stateName)
fullyConnectedLayer(400, ‘Name’, ‘CriticHiddenLayer1’)
reluLayer(‘Name’, ‘CriticReLU1’)
fullyConnectedLayer(300, ‘Name’, ‘CriticHiddenLayer2’)
reluLayer(‘Name’, ‘CriticReLU2’)
fullyConnectedLayer(1, ‘Name’, ‘CriticOutputLayer’)
];
% Create Actor Representation
actorOpts = rlRepresentationOptions(‘LearnRate’, actorLearningRate);
actor = rlDeterministicActorRepresentation(actorNetwork, stateSpec, actionSpec, actorOpts);% Create Environment
env = MYEnv();
% Define State and Action Specifications
stateSpec = env.getObservationInfo();
actionSpec = env.getActionInfo();
stateName = stateSpec.Name;
% Create Actor Network
actorNetwork = [
featureInputLayer(stateSpec.Dimension(1), ‘Name’, stateName)
fullyConnectedLayer(400, ‘Name’, ‘ActorHiddenLayer1’)
reluLayer(‘Name’, ‘ActorReLU1’)
fullyConnectedLayer(300, ‘Name’, ‘ActorHiddenLayer2’)
reluLayer(‘Name’, ‘ActorReLU2’)
fullyConnectedLayer(actionSpec.Dimension(1), ‘Name’, ‘ActorOutputLayer’)
tanhLayer(‘Name’, ‘ActorTanh’)
];
% Create Critic Network
criticNetwork = [
featureInputLayer(stateSpec.Dimension(1), ‘Name’, stateName)
fullyConnectedLayer(400, ‘Name’, ‘CriticHiddenLayer1’)
reluLayer(‘Name’, ‘CriticReLU1’)
fullyConnectedLayer(300, ‘Name’, ‘CriticHiddenLayer2’)
reluLayer(‘Name’, ‘CriticReLU2’)
fullyConnectedLayer(1, ‘Name’, ‘CriticOutputLayer’)
];
% Create Actor Representation
actorOpts = rlRepresentationOptions(‘LearnRate’, actorLearningRate);
actor = rlDeterministicActorRepresentation(actorNetwork, stateSpec, actionSpec, actorOpts); % Create Environment
env = MYEnv();
% Define State and Action Specifications
stateSpec = env.getObservationInfo();
actionSpec = env.getActionInfo();
stateName = stateSpec.Name;
% Create Actor Network
actorNetwork = [
featureInputLayer(stateSpec.Dimension(1), ‘Name’, stateName)
fullyConnectedLayer(400, ‘Name’, ‘ActorHiddenLayer1’)
reluLayer(‘Name’, ‘ActorReLU1’)
fullyConnectedLayer(300, ‘Name’, ‘ActorHiddenLayer2’)
reluLayer(‘Name’, ‘ActorReLU2’)
fullyConnectedLayer(actionSpec.Dimension(1), ‘Name’, ‘ActorOutputLayer’)
tanhLayer(‘Name’, ‘ActorTanh’)
];
% Create Critic Network
criticNetwork = [
featureInputLayer(stateSpec.Dimension(1), ‘Name’, stateName)
fullyConnectedLayer(400, ‘Name’, ‘CriticHiddenLayer1’)
reluLayer(‘Name’, ‘CriticReLU1’)
fullyConnectedLayer(300, ‘Name’, ‘CriticHiddenLayer2’)
reluLayer(‘Name’, ‘CriticReLU2’)
fullyConnectedLayer(1, ‘Name’, ‘CriticOutputLayer’)
];
% Create Actor Representation
actorOpts = rlRepresentationOptions(‘LearnRate’, actorLearningRate);
actor = rlDeterministicActorRepresentation(actorNetwork, stateSpec, actionSpec, actorOpts); observation names, rldeterministicactorrepresentation MATLAB Answers — New Questions

​