Low-code framework for building custom LLMs, neural networks, and other AI models
Full Changelog: https://github.com/ludwig-ai/ludwig/compare/v0.10.2...v0.10.3
true
in config (can be used in conjunction): adapter:
type: lora
use_rslora: false
use_dora: false
trainer:
eval_batch_size: auto
from_checkpoint=True
to LudwigModel.load()
:LudwigModel.load(model_dir, from_checkpoint=True)
Full Changelog: https://github.com/ludwig-ai/ludwig/compare/v0.10.1...v0.10.2
Full Changelog: https://github.com/ludwig-ai/ludwig/compare/v0.10.0...v0.10.1
get_quantization
by @jeffkinnison in https://github.com/ludwig-ai/ludwig/pull/3926
use_reentrant
to True
to fix Mixtral-7b
bug by @geoffreyangus in https://github.com/ludwig-ai/ludwig/pull/3928
Full Changelog: https://github.com/ludwig-ai/ludwig/compare/v0.9.3...v0.10.0
microsoft/phi-2
by @arnavgarg1 in https://github.com/ludwig-ai/ludwig/pull/3880
LLMEncoder
output to torch.float32
, freeze final layer at init. by @jeffkinnison in https://github.com/ludwig-ai/ludwig/pull/3900
LLMEncoder
by @jeffkinnison in https://github.com/ludwig-ai/ludwig/pull/3902
Full Changelog: https://github.com/ludwig-ai/ludwig/compare/v0.9.2...v0.9.3
name
and description
classmethods to IA3Config
by @jeffkinnison in https://github.com/ludwig-ai/ludwig/pull/3844
prepare_for_trianing
logic to ECD model for LLM encoder adapter initialization by @jeffkinnison in https://github.com/ludwig-ai/ludwig/pull/3874
libsox-dev
and update API by @arnavgarg1 in https://github.com/ludwig-ai/ludwig/pull/3879
Full Changelog: https://github.com/ludwig-ai/ludwig/compare/v0.9.1...v0.9.2
name
and description
classmethods to IA3Config
by @jeffkinnison in https://github.com/ludwig-ai/ludwig/pull/3844
Full Changelog: https://github.com/ludwig-ai/ludwig/compare/v0.9...v0.9.1
combiner_registry
to combiner_config_registry
, update decorator name by @ksbrar in https://github.com/ludwig-ai/ludwig/pull/3516
response
column for text output features during postprocessing by @arnavgarg1 in https://github.com/ludwig-ai/ludwig/pull/3521
datetime.date
date features by @jeffkinnison in https://github.com/ludwig-ai/ludwig/pull/3534
effective_batch_size
to auto-adjust gradient accumulation by @tgaddair in https://github.com/ludwig-ai/ludwig/pull/3533
LambdaLR
, and add cosine annealing LR scheduler as a decay method. by @justinxzhao in https://github.com/ludwig-ai/ludwig/pull/3555
prompt
validation check by @ksbrar in https://github.com/ludwig-ai/ludwig/pull/3564
check_module_parameters_updated
by @jeffkinnison in https://github.com/ludwig-ai/ludwig/pull/3567
4.33.2
by @arnavgarg1 in https://github.com/ludwig-ai/ludwig/pull/3637
model.predict
by @geoffreyangus in https://github.com/ludwig-ai/ludwig/pull/3666
max_new_tokens
based on output feature length, GMSL and model window size by @arnavgarg1 in https://github.com/ludwig-ai/ludwig/pull/3713
_initialize_llm
before moving to cuda. by @justinxzhao in https://github.com/ludwig-ai/ludwig/pull/3762
huggingface-hub>=0.19.0
by @brightsparc in https://github.com/ludwig-ai/ludwig/pull/3834
Full Changelog: https://github.com/ludwig-ai/ludwig/compare/v0.8...v0.9
max_new_tokens
based on output feature length, GMSL and model window size by @arnavgarg1 in https://github.com/ludwig-ai/ludwig/pull/3713
Full Changelog: https://github.com/ludwig-ai/ludwig/compare/v0.8.5...v0.8.6
model.predict
by @geoffreyangus in https://github.com/ludwig-ai/ludwig/pull/3666
Full Changelog: https://github.com/ludwig-ai/ludwig/compare/v0.8.4...v0.8.5