feat(demo): do not log search

feat: vmap optimizer (#166)

feat(demo): use fixed commit

doc: reference dalle playground

feat: scan layers + gradient checkpointing (#161)

Merge pull request #162 from borisdayma/demo-improvements

Merge branch 'main' of https://github.com/borisdayma/dalle-mini into main

Style: run isort.

Style: run black.

Display info about model and run, below predictions.

Update year: 2021-2022

Log dates

Move client api to backend.py

feat: better multi-node support (#158)

doc: link to new demo

feat(demo): high load

feat(demo): update space

feat(text): support emojis (#154)

feat: update shampoo

fix: smelu

fix: sinkformer

fix: support smelu

fix: einops is required

feat: allow relative position (#156)

feat: sinkhorn in lse mode (#155)

feat(demo): update model

fix: sinkformer gradient

feat(model): allow bias (#152)

feat(train): google-cloud-storage is optional

feat(train): rename logged config

feat: add sinkformer + custom final ln + pre-ln (#151)

feat: placeholders for more config

feat: add mini_glu config

feat: force final ln in encoder

feat: allow more configurations

fix: DeepNet doesn't scale weights of embedding/output layers (#150)

feat: remove unecessary LN

feat: update mini config

feat: add cogview

fix(textnormalizer): consider utf8 on windows (#148)

feat: implement transformer variants (#144)

feat(train): log norm and histograms (#143)

feat(data): super conditioning (#141)

feat: support pod (#139)

fix: no gradient checkpointing for new model

feat: no gradient checkpointing for params init

feat: update configs

feat(dev): require datasets

fix(train): consider schedule offset

feat(train): local jax cache