Building an MLP model to accurately classify the Fashion-MNIST images using the free version of ChatGPT
ChatGPT’s premium version has a code interpreter, but it doesn’t support the deep learning libraries such as Keras and TensorFlow required to execute the code. Hence, for this chapter, we will stick to the free version of ChatGPT.
Feature 1: Building the baseline model
Let’s craft our initial prompt for baseline model.
[Prompt]
I want to create a simple classification model for the Fashion-MNIST dataset (TAG 1.1) consisting of a single-layer MLP model (PIC 2.2). It should consist of the following steps (TAG1.2), provide one step at a time, and wait for the user’s feedback (PIC 2.2).
- Data Preprocessing: Normalize pixel values, flatten images into vectors, and encode categorical labels.
- Data Splitting: Partition the dataset into training, validation, and testing sets.
- Model Selection: Opt for a Multi-Layer Perceptron...
