dalle-mini / README.md
osanseviero's picture
Update README.md
91ed37e
|
raw
history blame
3.06 kB
metadata
language:
  - en
  - text-to-image
inference: false

DALL路E mini - Generate images from text

Model Description

This is an attempt to replicate OpenAI's DALL路E, a model capable of generating arbitrary images from a text prompt that describes the desired result.

DALL路E mini demo screenshot

This model's architecture is a simplification of the original, and leverages previous open source efforts and available pre-trained models. Results have lower quality than OpenAI's, but the model can be trained and used on less demanding hardware. Our training was performed on a single TPU v3-8 for a few days.

Components of the Architecture

The system relies on the Flax/JAX infrastructure, which are ideal for TPU training. TPUs are not required, both Flax and JAX run very efficiently on GPU backends.

The main components of the architecture include:

  • An encoder, based on BART. The encoder transforms a sequence of input text tokens to a sequence of image tokens. The input tokens are extracted from the text prompt by using the model's tokenizer. The image tokens are a fixed-length sequence, and they represent indices in a VQGAN-based pre-trained codebook.

  • A decoder, which converts the image tokens to image pixels. As mentioned above, the decoder is based on a VQGAN model.

The model definition we use for the encoder can be downloaded from our Github repo. The encoder is represented by the class CustomFlaxBartForConditionalGeneration.

To use the decoder, you need to follow the instructions in our accompanying VQGAN model in the hub, flax-community/vqgan_f16_16384.

How to Use

The easiest way to get familiar with the code and the models is to follow the inference notebook we provide in our github repo. For your convenience, you can open it in Google Colaboratory: Open In Colab

If you just want to test the trained model and see what it comes up with, please visit our demo, available in 馃 Spaces.

Additional Details

Our report contains more details about how the model was trained and shows many examples that demonstrate its capabilities.