Labeling is assigning categories to your images to create examples hat teach Lobe. These examples are commonly known as a dataset. From this dataset, Lobe will learn to automatically predict a label for a given image.

Import and label your images in Lobe.

• Images - import common image files directly from your computer. Lobe supports JPEG, PNG, BMP, and WebP formats.
• Camera - use any connected camera source to capture images directly in Lobe. You can optionally provide a label for these images. Hold down the camera button to capture a burst of images.

Import an existing dataset.

• Folders - import existing labeled images by using folder names as the labels.

You can create new labels or edit existing ones by using the text box in the bottom corner of each image.

Note

• The max image size Lobe can process is 178,956,970 pixels. For a square image, that is about 13,300 x 13,300 pixels. We recommend staying lower-resolution for faster processing because Lobe will resize and crop your image to a 224 x 224 square.

Training is how your model learns to predict the correct labels from your examples. You can think of your examples as a collection of flashcards. During training, your model will continually look through the flashcards and try to find similar patterns that help it guess the right answers.

Read moremachine learning basics

Read more aboutlabeling your examples

Lobe automatically starts training when your examples meet the minimum requirements. To start training, you need:

• Imported images tolabel as examples
• At least two labels
• At least five images per label

Lobe will also follow best practices to continue training when you make changes to your examples. If you make large changes or add/remove labels, Lobe will reset training and start training a new model.

After automatic training has completed, you can manually optimize your model to train for longer for better real-world performance (File > Optimize Model).

Training time is quite variable and depends on how hard it is to distinguish between the labels in your examples and how many examples you have. It can take anywhere from minutes to hours, and sometimes days for very large problems.

You can hover over the training progress to see a time estimate. This training time estimate is updated every few seconds based on your progress and computer’s current processing speed. You may see it fluctuate if you are performing other tasks on your computer as available CPU and memory change.

Your results show you which images your model is predicting correctly and incorrectly. Correct predictions have green labels and incorrect predictions have red labels. The width of the label bar represents how confident the model was in that prediction.

Hovering over a predicted label will show the true label you gave that image. The more correct the predictions, the better the model is performing.

You can view and sort your images in different ways to check:

• Whether your model is successfully learning all the labels with iew > All Images selected.
• Approximately how well it will work on new images with View > Test Images selected. Learn more about test images.
• Which images and labels confuse your model by selecting your individual labels in the sidebar.

Select your individual labels in the sidebar to see results specific to that label. Lobe will show you which other labels are commonly confused with this one.

Try to spot what similarities exist between the images that confuse the model and the images in the true label. For example, you may notice the same background color across the confused images and the images in your label. If you notice these patterns, try to collect varying images that look like the confused image to better teach Lobe what patterns to ignore. In this example, collect more images in your selected label with varying background colors.

To see where your model is most confused, look at the most confident incorrect predictions and the least confident correct predictions.

• Select View > Correct First to view your predictions sorted by most confident to least confident. Try to find any patterns where your model was very confident in its wrong prediction.
• Select View > Incorrect First to view your predictions sorted by least confident to most confident. See where your model is least confident with its correct predictions.

Collect more variations of images that have similar patterns to these two cases to help your model improve.

Use images from your computer or use the webcam as a video feed to test your model on new images live. Try to actively trick your model and see patterns where it is getting images incorrect. Help improve you model by giving feedback on its predictions.

To give your model feedback, you can edit the prediction text box directly and add the image and label as an example. Alternatively, click the correct or incorrect buttons on the image to add it as an example. Lobe will automatically train with these new examples to continually improve.

Currently, you can only view a single image at a time or use your webcam as a video feed.

Many factors can cause your model to not perform well when playing with new images. Check out Improving for tips on what to look for and how to improve your model.

Your model is a collection of files that other programs can load to run predictions. These files store both the structure of your model and the weights that are a result of training.

You can directly upload your model to Power Platform to use in Power Apps or Power Automate. You can also use the model files locally in your own app or in most major cloud platforms to create an API. Lobe also hosts your model as a local API to help kickstart your app development.

Lobe provides a few workflows for using models: no-code apps on Microsoft’s Power Platform, calling a local API, adapting starter projects, or working with model files directly.

We are continually expanding the ways you can use your model. Current recommendations:

Your examples are your model’s source of truth about the world. Here are some best practices for building successful projects:

• Understand your problem - start simple, expand over time. Break your scenario down into smaller experiments to prototype and then expand over time.
• Lobe only learns from the examples you import. Try to collect examples that cover the different types of images your model will see and make predictions on in the future.
• More images always help - new and unique images are better. The more unique and different the images are, the better your model will learn to generalize.
• If you can’t classify the label from looking at an image, it will also be difficult for Lobe. Make the image content as large and relevant to your label as possible.

Compare the images in Use with the images in your examples:

• Your model may have memorized your examples instead of learning to generalize to new images. This is commonly called overfitting. Check Results to see if your model is overfitting.
• These new images are not represented by your examples. Try to play with edge cases and ‘trick’ Lobe as much as possible. Make sure your examples contain all these new variations and conditions seen from Use. You can add new images as examples directly from Use so that Lobe can continually improve.

Lobe will always predict one of your labels even if your image does not contain any related content. If you expect your model to see these types of images, create a ‘None’ label and add variations of these images as examples. You can use this ‘None’ label as a placeholder when waiting for relevant predictions.

• Your model may have memorized your examples instead of learning to generalize to new images. This is commonly called overfitting. Check Results to see if your model is overfitting.
• These new images are not represented by your examples. Try to play with edge cases and ‘trick’ Lobe as much as possible. Make sure your examples contain all these new variations and conditions seen from Use. You can add new images as examples directly from Use so that Lobe can continually improve.

Follow announcements on our website and social media:

• Website
• Twitter
• Facebook
• Reddit

Watch our intro tour for a tutorial.

Global

• Check for updates: ctrl/cmd + shift + U

Training is a very computation-heavy process. Lobe will use a lot of memory and CPU bandwidth to train as quickly as possible. This generally means your computer will have less CPU and memory available for other apps to use.