5 Things to Know Before Customizing Your First Machine Learning Model with MediaPipe Model Maker

1. Preparing your data for training will look different depending on the type of model you're customizing. Be sure to make the time to properly source and annotate your data, and make sure you have the right to use any images or text you include.

2. Simplify your custom model by focusing on a specific, achievable goal. Training for edge cases can introduce errors or strange behavior, so keep it simple and focused.

3. Start with a small dataset of at least 100 examples. You may not need more than 100 if there aren't many possible iterations of the data, but you may need more if you need to show the possible iterations of the data.

4. Use MediaPipe Studio to try out different options and see results with low latency. This can save time during the development process and make it easier to fine-tune your model.

5. Remember that building a machine learning model is a challenging process and it may take time to achieve your desired results. Keep experimenting and iterating, and don't be afraid to ask for help or seek out additional resources if needed.