MLProjects

1. MimoPracticesML
   • catdogimage: following model was deployed through teachable machines/ml5.js
   • Neural Network
   • Transfer Learning
   • Pre Trained model
   • Rock paper scissor

3 ways ml can be incorporated in our projects:

• pre-trained model (mobilenet, mobilenetv2, mobilenetv3)
• transfer learning ( using knowledge of one task to improve performance of another)
• custom models ( own model from scratch)

Basic ML Concepts:

• Loss function: function used to calc loss value.
• Loss value: way to determine margine of error from guess answer to actual answer, lesser the loss value more are we accurate.
• Epoch: one complete pass through the training dataset.
• Batch size: no of training examples used in one iteration.