Commit History

Update `resume_from_checkpoint` to use `from_pretrained`.
bb3f53e

Pedro Cuenca commited on

Never consider local dirs as remote wandb references.
08dd098

Pedro Cuenca commited on

Load tokenizer associated to the model checkpoint, if possible.
a77c0d4

Pedro Cuenca commited on

Store resolved path after loading model.
55a631d

Pedro Cuenca commited on

Use model configuration unless a specific one is supplied.
5ec61cc

Pedro Cuenca commited on

Override from_pretrained to support wandb artifacts.
1023afa

Pedro Cuenca commited on

feat: support pypi
f5dba1e

boris commited on

doc: update contributions
e3b1b56

boris commited on

Merge pull request #117 from borisdayma/fix-inference
ef985be
unverified

boris commited on

fix(inference): use float32 + flatten logits
71c4de3

boris commited on

Merge pull request #115 from borisdayma/feat-shampoo
3a3d375
unverified

boris commited on

feat: update inference pipeline
af807f7

boris commited on

style: isort
531cd78

boris commited on

style: apply to distributed_shampoo
e669c1b

boris commited on

feat: add best_effort_memory_usage_reduction
4d518c7

boris commited on

doc: add reference to Distributed Shampoo
db882b8

boris commited on

feat: update distributed_shampoo
b90198c

boris commited on

fix: weight decay Adam + speed logging
7143593

boris commited on

feat: add micro config
e501f71

boris commited on

fix: shampoo -> distributed shampoo
edae62d

boris commited on

feat: update params
604a65d

boris commited on

feat: add shampoo optimizer
0b87452

boris commited on

feat: update sweep
e1555d4

boris commited on

feat: create config files
dc5c024

boris commited on

feat: allow abstract_init
772415c

boris commited on

fix: typo
5c84978

boris commited on

feat: log more metrics
1b757dc

boris commited on

feat: shard by host is optional
901ff72

boris commited on

feat: load data first
fdf7698

boris commited on

feat: display local TPU's
15993e3

boris commited on

fix: check local TPU instances only
87fed1b

boris commited on

feat: add config
3f13951

boris commited on

feat(setup): require braceexpand for dev
98f1db7

boris commited on

fix(data): type
c6ebb14

boris commited on

feat(train): handle multi-hosts
5b533b5

boris commited on

feat: split shards by host
ed93c8a

boris commited on

style
a6252c9

boris commited on

feat(data): accept braceexpand notation
5ee6e60

boris commited on

feat: minor improvements
53dade7

boris commited on

fix(config): set min/max for generation
eb24dbc

boris commited on

fix: update model name
61c93f2

boris commited on

fix(train): update model name
b257ca8

boris commited on

feat: log num_params
1f57ad7

boris commited on

fix: causal_mask based on image tokens
8654dc9

boris commited on

refactor(model): inherit from HF Flax & simplify
972bc8d

boris commited on

style
6f1f2d9

boris commited on

fix: adjust training script + dataloader
a96f4dc

boris commited on

fix(model): use correct params
a11892f

boris commited on

Merge branch 'main' of https://github.com/borisdayma/dalle-mini into add-custom-model
f234ccf

boris commited on

feat(requirements): add pillow
26651dd

boris commited on