Image Classification(Deep Learning) using C# ML .NET

[@sureshpal5569]

great video , can u share the code , please

[@nguyenthanhphong5513]

Thank you so much, I have already resolved my problem.

[@mohammedmuzzammil1184]

How to make this from camera input, thanks a lot.

[@BiswaRanjan]

That's a good question. I will try and update you.

[@kemmunvlogs7388]

@@BiswaRanjan did you make a video about this one, thanks a lot?

[@dr.khushbooagarwal786]

Can you please make a tutorial on the auto generated code and what algorithm is used by it

[@georgealbo]

Can you please tell.how to train a model in c sharp exactly as the model.builder and create model.zip file thank you for this.video

[@BiswaRanjan]

Hi, I will a create video on coming sessions. Please stay tuned, enjoy and learn. Thank you for watching.

[@georgealbo]

Thanks

[@georgealbo]

@@BiswaRanjan also loading and saving a trained Model to be consumed. iam waiting thanks in advance

[@georgealbo]

Thank.you iam asking how to do.this using programming I mean make a training button instead of wizard

[@Eefficienzaartificiale-wl2su]

Hi! nice video! a question: why does it take so long for the prediction if the model is already loaded? it should be fast. it's slow for me too.

[@BiswaRanjan]

Hi, Thanks for watching. If we run in our local machine, it takes times as per system configuration.

[@okanyosun]

how many time does it take 30k and 42mb data on a computer have 16gb ram and ryzen 5 2600 cpu ? should i need azure ?

[@nikhilshinde6195]

Hey, thanks for the tutorial. I'm getting error while I training the dataset. The error is : File not found. at System.Diagnostics.Process.StartWithCreateProcess(ProcessStartInfo startInfo) Please help me

[@georgealbo]

Dear sit how can i program this model in c# instead of using Model builder

[@BiswaRanjan]

When the console app creating, There is modelbuilder.cs file, in there please check for Train() method. you can use like wise

[@georgealbo]

@@BiswaRanjan actually i created the Model in 2022 it did not create ModelBuilder , iam waiting ... also iam wishing if you save the Model and load it to be used for consumption . thanks in advace

[@sureshpal5569]

SOURCE CODE FILE IS NOT AVAILABLE

[@desifood1895]

Hi sir, please help me to find the source code ,that link shows that file not found....

[@BiswaRanjan]

Hi curry , unfortunately due low space I deleted source code. Very soon I am going to create same video from .Net core 6 web.API, I will update you over that.

[@vkeychaudhari]

Can you share the source code?

[@timMathsHelp]

Thank you, very helpful

[@BiswaRanjan]

Glad to hear that!

[@sanketmetkar]

while training the data it shows TensorFlow exception triggered while loading model how to solve it?

[@BiswaRanjan]

Could you please send the exact exception message?

[@sanketmetkar]

@@BiswaRanjan Tensorflow exception triggered while loading model. at System.Runtime.ExceptionServices.ExceptionDispatchInfo.Throw() at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task) at Microsoft.ML.ModelBuilder.AutoMLService.Experiments.AutoMLExperiment`3.d__21.MoveNext() in /_/src/Microsoft.ML.ModelBuilder.AutoMLService/Experiments/AutoMLExperiment.cs:line 108 --- End of stack trace from previous location where exception was thrown --- at System.Runtime.ExceptionServices.ExceptionDispatchInfo.Throw() at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task) at Microsoft.ML.ModelBuilder.AutoMLEngine.d__30.MoveNext() in /_/src/Microsoft.ML.ModelBuilder.AutoMLService/AutoMLEngineService/AutoMLEngine.cs:line 147

[@BiswaRanjan]

​@@sanketmetkar Sorry for late reply. My response:- When a system facing performance issue like the below scenario : 1- Machine configuration is slow. 2- Trying to train the model with large size of the image, when the system is slow. I suggest trying to close all your application from the machine and first try with a small amount of image. If is working or not. If it working then again train the model by putting all images in the specified folder. Try to update Visual studio 2019 16.7.

[@sanketmetkar]

@@BiswaRanjan i tried all the ways but still same error. don't know what is happening

[@BiswaRanjan]

@@sanketmetkar Please try to use the below pipeline and check again. var trainingPipeline = mapLabelTransform .Append(mlContext.Model.ImageClassification( "ImagePath", "LabelAsKey", arch: ImageClassificationEstimator.Architecture.ResnetV2101, epoch: 100, batchSize: 30, metricsCallback: (metrics) => Console.WriteLine(metrics)));

[@傻东西]

Hi, may i know where can i get the full codes?

[@BiswaRanjan]

Hi, Please download the source code at : easyupload.io/i336ea

[@sukeshsubaharan4324]

Hi, I am continuously getting an error during runtime that says Microsoft.ML.Data could not load one of its dependencies. I tried changing the .NET version, but nothing works. Could you please suggest a fix? PS. I followed your tutorial to the letter

[@BiswaRanjan]

Hi, Thanks for watching. Could you please copy paste the exact error/exception messages here. There might be chance, you need to update visual studio and model builder.

[@sukeshsubaharan4324]

@@BiswaRanjan I am running VS 2019 Community, v16.7.6. I updated it yesterday. The error I am getting: System.IO.FileNotFoundException: Microsoft.ML.Data, Version=1.0.0.0, Culture=neutral, PublicKeyToken=cc7b13ffcd2ddd51. Could not load file or one of its dependencies

[@sukeshsubaharan4324]

@@BiswaRanjan that error is referenced to the output prediction line in the code btw

[@BiswaRanjan]

@@sukeshsubaharan4324 Please confirm, you are using .Net Core not .Net Framework. If you are using .Net Core then please check below. Could you check if Microsoft.ML is installed correctly, Seems that it can't load the right assembly. Model Builder should install Microsoft.ML automatically in generated console app and Model project. If possible, could you create a new project and start a new train to see if the DLL is not found exception happened again? And also check if Microsoft.ML installed correctly in the generated console app and Model project? If still the issue is persists you need to reinstall the VS2019.

[@sukeshsubaharan4324]

@@BiswaRanjan thank you! I will try this and let you know.

[@NarunartKudwongkeaw]

When I have a new image or new data, how to update for this project ?

[@BiswaRanjan]

In the Added ConsoleApp application in your solutions, please look for the ModelBuilder.cs -> public static ITransformer TrainModel() you will get the ideas to train your model after new input. Thank you for watching.

[@CarlosMachado-ub7rl]

Some errors after building: CS0006 Metadata file 'D:\PROGRAMMING\CSharp\MLProject\ImageClassify\ImageClassifyML.Model\bin\Debug etstandard2.0\ImageClassifyML.Model.dll' could not be found ImageClassify D:\PROGRAMMING\CSharp\MLProject\ImageClassify\ImageClassify\CSC 1 Active

[@mohammedmuzzammil1184]

why this result.Prediction is null

[@shubhamberde-fg4ez]

👍👍👍👍

[@BiswaRanjan]

Source code available for download at easyupload.io/i336ea

[@ZNAXNOR]

FILE NOT FOUND error

[@samuelgirmagirma4019]

C# not she#