Перейти к содержимому
Меню
Русский

Сценарий использования

Image generation

Produce images from a text prompt, and understand why the two dominant approaches differ.

Посмотреть рекомендуемый стек с обоснованием и альтернативами →

Что требуется

В порядке внедрения. Каждый шаг открывается в графе знаний.

  1. 1 Generative model

    Diffusion

    technique

    Generates data by learning to reverse a gradual noising process.

  2. 2 Alternative approach

    GAN

    architecture

    Generator and discriminator trained against each other to synthesise realistic data.

  3. 3 Architecture

    CNN

    architecture

    Convolutional neural network, sharing weights across positions to exploit spatial structure.

  4. 4 Conditioning

    Attention

    concept

    Mechanism that weights each part of an input by its relevance to every other part.