How to specify my own path of *.meta when doing transfer learning?

[mailspeed]

System information

• Windows 10 Enterprise 64bit/19042.1165:
• .NET 5.0:

Issue

• I'm trying to build a offline transfer learning application using ML.NET
• When creating "ImageClassificationTrainer.Options()", I set the Arch parameter(e.g. MobilenetV2). Then coming to trainingPipeline.Fit(), ML.NET will automatically download the "mobilenet_v2.meta" to "C:\Users\user\AppData\Local\Temp\MLNET\"
• What I expect: Is it possible to specify a path of "mobilenet_v2.meta" to any local disk path that I wished (for example I've pre-downloaded mobilenet_v2.meta some where else) instead of the automatically downloaded path. You know when deploy my application to production, the target PC will probabaly have no internet connection, I have to attach mobilenet_v2.meta file to the application package.

Source code / logs

using System; using System.Collections.Generic; using System.Linq; using System.IO; using Microsoft.ML; using static Microsoft.ML.DataOperationsCatalog; using Microsoft.ML.Vision;

namespace DeepLearning_ImageClassification { class Program { static void Main(string[] args) { var projectDirectory = Path.GetFullPath(Path.Combine(AppContext.BaseDirectory, "../../../")); var workspaceRelativePath = Path.Combine(projectDirectory, "workspace"); var assetsRelativePath = Path.Combine(projectDirectory, "assets");

        MLContext mlContext = new MLContext();

        // You must unzip assets.zip before training
        IEnumerable<ImageData> images = LoadImagesFromDirectory(folder: assetsRelativePath, useFolderNameAsLabel: true);

        IDataView imageData = mlContext.Data.LoadFromEnumerable(images);

        IDataView shuffledData = mlContext.Data.ShuffleRows(imageData);

        var preprocessingPipeline = mlContext.Transforms.Conversion.MapValueToKey(
                inputColumnName: "Label",
                outputColumnName: "LabelAsKey")
            .Append(mlContext.Transforms.LoadRawImageBytes(
                outputColumnName: "Image",
                imageFolder: assetsRelativePath,
                inputColumnName: "ImagePath"));

        IDataView preProcessedData = preprocessingPipeline
                            .Fit(shuffledData)
                            .Transform(shuffledData);

        TrainTestData trainSplit = mlContext.Data.TrainTestSplit(data: preProcessedData, testFraction: 0.3);
        TrainTestData validationTestSplit = mlContext.Data.TrainTestSplit(trainSplit.TestSet);

        IDataView trainSet = trainSplit.TrainSet;
        IDataView validationSet = validationTestSplit.TrainSet;
        IDataView testSet = validationTestSplit.TestSet;

        var classifierOptions = new ImageClassificationTrainer.Options()
        {
            FeatureColumnName = "Image",
            LabelColumnName = "LabelAsKey",
            ValidationSet = validationSet,
            Arch = ImageClassificationTrainer.Architecture.MobilenetV2,
            MetricsCallback = (metrics) => Console.WriteLine(metrics),
            TestOnTrainSet = false,
            ReuseTrainSetBottleneckCachedValues = true,
            ReuseValidationSetBottleneckCachedValues = true
        };

        var trainingPipeline = mlContext.MulticlassClassification.Trainers.ImageClassification(classifierOptions)
            .Append(mlContext.Transforms.Conversion.MapKeyToValue("PredictedLabel"));

        ITransformer trainedModel = trainingPipeline.Fit(trainSet);

        ClassifySingleImage(mlContext, testSet, trainedModel);

        ClassifyImages(mlContext, testSet, trainedModel);

        Console.ReadKey();
    }

    public static void ClassifySingleImage(MLContext mlContext, IDataView data, ITransformer trainedModel)
    {
        PredictionEngine<ModelInput, ModelOutput> predictionEngine = mlContext.Model.CreatePredictionEngine<ModelInput, ModelOutput>(trainedModel);

        ModelInput image = mlContext.Data.CreateEnumerable<ModelInput>(data,reuseRowObject:true).First();

        ModelOutput prediction = predictionEngine.Predict(image);

        Console.WriteLine("Classifying single image");
        OutputPrediction(prediction);
    }

    public static void ClassifyImages(MLContext mlContext, IDataView data, ITransformer trainedModel)
    {
        IDataView predictionData = trainedModel.Transform(data);

        IEnumerable<ModelOutput> predictions = mlContext.Data.CreateEnumerable<ModelOutput>(predictionData, reuseRowObject: true).Take(10);

        Console.WriteLine("Classifying multiple images");
        foreach (var prediction in predictions)
        {
            OutputPrediction(prediction);
        }
    }

    private static void OutputPrediction(ModelOutput prediction)
    {
        string imageName = Path.GetFileName(prediction.ImagePath);
        Console.WriteLine($"Image: {imageName} | Actual Value: {prediction.Label} | Predicted Value: {prediction.PredictedLabel}");
    }

    public static IEnumerable<ImageData> LoadImagesFromDirectory(string folder, bool useFolderNameAsLabel = true)
    {
        var files = Directory.GetFiles(folder, "*",
            searchOption: SearchOption.AllDirectories);

        foreach (var file in files)
        {
            if ((Path.GetExtension(file) != ".jpg") && (Path.GetExtension(file) != ".png"))
                continue;

            var label = Path.GetFileName(file);

            if (useFolderNameAsLabel)
                label = Directory.GetParent(file).Name;
            else
            {
                for (int index = 0; index < label.Length; index++)
                {
                    if (!char.IsLetter(label[index]))
                    {
                        label = label.Substring(0, index);
                        break;
                    }
                }
            }

            yield return new ImageData()
            {
                ImagePath = file,
                Label = label
            };
        }
    }
}

class ImageData
{
    public string ImagePath { get; set; }

    public string Label { get; set; }
}

class ModelInput
{
    public byte[] Image { get; set; }

    public UInt32 LabelAsKey { get; set; }

    public string ImagePath { get; set; }

    public string Label { get; set; }
}

class ModelOutput
{
    public string ImagePath { get; set; }

    public string Label { get; set; }

    public string PredictedLabel { get; set; }
}

}

[luisquintanilla]

Hi @mailspeed

You should be able to customize the path by setting the WorkspacePath in your classifierOptions

https://docs.microsoft.com/dotnet/api/microsoft.ml.vision.imageclassificationtrainer.options?view=ml-dotnet

[mailspeed]

Thanks a lot for your reply. I'll try that.

Get Outlook for Androidhttps://aka.ms/AAb9ysg

---

From: Luis Quintanilla @.> Sent: Thursday, September 16, 2021 10:57:55 PM To: dotnet/machinelearning @.> Cc: mailspeed @.>; Mention @.> Subject: Re: [dotnet/machinelearning] How to specify my own path of *.meta when doing transfer learning? (#5927)

Hi @mailspeedhttps://github.com/mailspeed

You should be able to customize the path by setting the WorkspacePath in your classifierOptions

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/dotnet/machinelearning/issues/5927#issuecomment-920978212, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AEOY3YOL7QWLHWQLMTR6PQLUCIAXHANCNFSM5DV53YUQ. Triage notifications on the go with GitHub Mobile for iOShttps://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Androidhttps://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[michaelgsharp]

@mailspeed I'm going to close this for now as it seems @luisquintanilla was able to help. If you still have more questions feel free to either re-open or create a new issue.