No LoadData method for image classification  #2440

Activity

1. beccamc

   added

   Bug BashGeneral MB bashGeneral MB bash

   on Jan 11, 2023
2. beccamc

   changed the title [-]No LoadData method for Image[/-] [+]No LoadData method for image classification [/+] on Jan 11, 2023
3. zewditu commented on Jan 12, 2023

   

   zewditu

   on Jan 12, 2023

   Last edited by zewditu

   Contributor

   More actions

   Does the image has a loadData Method? does this include the template or discuss load method for image
4. beccamc commented on Jan 12, 2023

   

   beccamc

   on Jan 12, 2023

   ContributorAuthor

   More actions

   Nope, I need to provide the source code!
5. luisquintanilla

   added this to the February 2023 milestone on Jan 12, 2023
6. luisquintanilla

   assigned

   beccamc

   on Jan 12, 2023
7. beccamc

   removed

   Bug BashGeneral MB bashGeneral MB bash

   on Jan 12, 2023
8. beccamc

   modified the milestones: February 2023, March 2023 on Feb 1, 2023
9. github-actions

   added

   Stale

   on Mar 4, 2023
10. luisquintanilla

    assigned

    zewditu

    and unassigned

    beccamc

    on Mar 16, 2023
11. LittleLittleCloud commented on Mar 23, 2023

    

    LittleLittleCloud

    on Mar 23, 2023

    Last edited by LittleLittleCloud

    Member

    More actions

    Here's an example of loading images from folder.

    Example for local image classification

    public static IDataView LoadImagesFromFolder(MLContext context, string folder)
    {
        var res = new List<ModelInput>();
        var allowedImageExtensions = new[] { ".png", ".jpg", ".jpeg", ".gif" };
        DirectoryInfo rootDirectoryInfo = new DirectoryInfo(folder);
        DirectoryInfo[] subDirectories = rootDirectoryInfo.GetDirectories();
    
        if (subDirectories.Length == 0)
        {
            throw new Exception("fail to find subdirectories");
        }
    
        foreach (DirectoryInfo directory in subDirectories)
        {
            var imageList = directory.EnumerateFiles().Where(f => allowedImageExtensions.Contains(f.Extension.ToLower()));
    
            if (imageList.Count() > 0)
            {
                res.AddRange(imageList.Select(i => new ModelInput 
                {
                    Label = directory.Name,
                    ImageSource = File.ReadAllBytes(i.FullName),
                }));
            }
        }
    
        return context.Data.LoadFromEnumerable(res);
    }
    
    ## Example for azure image classification
    
    ```csharp
    public static IDataView LoadImagesFromFolder(MLContext context, string folder)
    {
        var res = new List<ModelInput>();
        var allowedImageExtensions = new[] { ".png", ".jpg", ".jpeg", ".gif" };
        DirectoryInfo rootDirectoryInfo = new DirectoryInfo(folder);
        DirectoryInfo[] subDirectories = rootDirectoryInfo.GetDirectories();
    
        if (subDirectories.Length == 0)
        {
            throw new Exception("fail to find subdirectories");
        }
    
        foreach (DirectoryInfo directory in subDirectories)
        {
            var imageList = directory.EnumerateFiles().Where(f => allowedImageExtensions.Contains(f.Extension.ToLower()));
    
            if (imageList.Count() > 0)
            {
                res.AddRange(imageList.Select(i => new ModelInput 
                {
                    Label = directory.Name,
                    ImageSource = MLImage.LoadFromFile(i.FullName), // this is the only difference comparing to local image training
                }));
            }
        }
    
        return context.Data.LoadFromEnumerable(res);
    }

    ❤️React with ❤️1Reacted by Becca McHenry
12. github-actions

    removed

    Stale

    on Mar 24, 2023
13. zewditu commented on Mar 24, 2023

    

    zewditu

    on Mar 24, 2023

    Contributor

    More actions

    for Azure, it should be CreateFromFile instead of LoadFromFile.
    Is that right ?@LittleLittleCloud
14. github-actions

    added

    Stale

    on Apr 24, 2023