Redco release v0.4.23

• An argument updated in Trainer.fit(): save_every_ckpt -> save_ckpt_every_k_epochs
• Added params_sharded and opt_state_sharded in Trainer.__init__(), for memory saving.

Redco release v0.4.22

• Simplified argument names for the random key in loss_fn() and pred_fn():
  • train_rng/pred_rng -> rng

redco-v0.4.21

• Accelerated Inference for multi-host, purely data parallel case
• Added optional argument train_step_fn in Trainer for fully customizing every training step, e.g., per-sample gradient noising for data-private training.
• Slight argument name change in Deployer.get_lr_schedule_fn(): warmup_rate -> warmup_ratio

redco-v0.4.20

• Updated data example type support -- can be a list of whatever types now, e.g., examples=[str, str, str, ...] or examples=[dict, dict, dict, ...]
• Updated mixed-precision training -- by setting compute_dtype, e.g., Trainer(compute_dtype=jnp.bfloat16).

Redco release v0.4.19

• Accelerated multi-host running
• Updated WandB login, e.g., redco.Deployer(wandb_init_kwargs={'project': '...', 'name': '...'})
• Updated customization of params_sharding_rules

Redco release v0.4.18

Simplified checkpoint loading.

Redco release v0.4.17

• Faster ckpt saving and loading
• Text-to-image example updated with model parallelism for StableDiffusion

Redco release v0.4.16

Redco version 0.4.16.

Redco release v0.4.15

Redco version 0.4.15. Updated ckpt saving & loading.

Redco release v0.4.14

Redco version 0.4.14. Updated SLURM setting. See language_modeling or text_to_text for detailed use cases.