Capability
Image Generation With Dall E Models And Size Quality Control
20 artifacts provide this capability.
Want a personalized recommendation?
Find the best match →Top Matches
1
DALLE-pytorchFramework45/100
via “evaluation metrics and generation quality assessment”
Implementation / replication of DALL-E, OpenAI's Text to Image Transformer, in Pytorch
Unique: Computes training metrics (reconstruction loss, language modeling loss) and optional perceptual metrics (LPIPS, FID). Supports periodic sample generation during training for visual quality assessment.
vs others: More complete than basic loss tracking; includes optional perceptual metrics and sample generation. Enables data-driven model selection vs manual inspection.