ludwig-ai/ludwig issues and pull requests

#4044 - [DO NOT MERGE] TESTING GITHUB ACTIONS PIPELINE FOR PYTHON UPDATE

Pull Request - State: open - Opened by ethanreidel 2 months ago - 1 comment

#4043 - pip install ludwig fails on PyYAML

Issue - State: open - Opened by eladrave 2 months ago

#4041 - Add Python 3.10, 3.11, 3.12 compatibility

Pull Request - State: open - Opened by mhabedank 3 months ago - 2 comments

#4040 - I think there is big problem when install ludwig, even I try many times with many python version

Issue - State: open - Opened by karamata 3 months ago - 6 comments

#4039 - [pre-commit.ci] pre-commit suggestions

Pull Request - State: open - Opened by pre-commit-ci[bot] 3 months ago

#4038 - WIP: Remove torchtext dependency

Pull Request - State: open - Opened by mhabedank 3 months ago - 1 comment

#4037 - Suggestion: transition from marshmallow to pydantic 2

Issue - State: open - Opened by ksbrar 3 months ago - 2 comments
Labels: dependency

#4036 - Ludwig not working on Kaggle

Issue - State: open - Opened by mhabedank 4 months ago

#4035 - Refactor BERTTokenizer

Issue - State: open - Opened by mhabedank 4 months ago - 1 comment
Labels: help wanted, dependency

#4034 - Refactor GPT2BPETokenizer

Issue - State: open - Opened by mhabedank 4 months ago
Labels: help wanted, dependency

#4033 - Refactor CLIPTokenizer

Issue - State: open - Opened by mhabedank 4 months ago
Labels: help wanted, dependency

#4032 - Refactor SentencePieceTokenizer

Issue - State: open - Opened by mhabedank 4 months ago
Labels: help wanted, dependency

#4031 - Refactor NgramTokenizer

Issue - State: open - Opened by mhabedank 4 months ago - 4 comments
Labels: help wanted, dependency

#4030 - Unable to Install Ludwig Package in Dataproc Cluster Jupyter Notebook

Issue - State: open - Opened by ayyappagundu 4 months ago - 3 comments
Labels: bug, dependency

#4029 - Ludwig installation not working any where

Issue - State: open - Opened by dexterslabs 4 months ago - 2 comments
Labels: bug, waiting for answer, dependency

#4028 - Update text_encoders.py

Pull Request - State: open - Opened by sileod 5 months ago - 1 comment

#4027 - Remove codeowners file

Pull Request - State: open - Opened by arnavgarg1 5 months ago

#4026 - Fix for issue 4013

Pull Request - State: closed - Opened by jhudsy 5 months ago

#4025 - Ludwig Installation Not Working

Issue - State: open - Opened by abedkhooli 5 months ago - 8 comments
Labels: bug, llm, dependency

#4024 - Fix typos

Pull Request - State: closed - Opened by omahs 6 months ago

#4022 - Fixed options dublicate in documentation for EmbeddingInitializerField

Pull Request - State: closed - Opened by mhabedank 6 months ago - 8 comments

#4021 - Update the trunk version to "0.10.4.dev"

Pull Request - State: closed - Opened by alexsherstinsky 6 months ago - 1 comment

#4020 - Update ludwig version to v0.10.4

Pull Request - State: closed - Opened by alexsherstinsky 6 months ago

#4019 - Update ludwig version to v0.10.4

Pull Request - State: closed - Opened by alexsherstinsky 6 months ago

#4018 - Update ludwig version to v0.10.4

Pull Request - State: closed - Opened by alexsherstinsky 6 months ago

#4017 - Errors opening DevContainer on Windows

Issue - State: open - Opened by CraigRichards 6 months ago - 1 comment
Labels: bug

#4016 - training_progress.json can be an invalid json

Issue - State: open - Opened by nqbao 7 months ago - 2 comments
Labels: bug, help wanted

#4015 - AttributeError: module 'torchtext' has no attribute 'transforms'

Issue - State: open - Opened by nqbao 7 months ago - 5 comments
Labels: bug, llm, dependency

#4014 - Add Phi-3 Support

Pull Request - State: closed - Opened by arnavgarg1 7 months ago - 4 comments

#4013 - Torchtext undefined module when using gpt2bpe tokenizer

Issue - State: open - Opened by jhudsy 8 months ago - 2 comments
Labels: bug, waiting for answer, llm, dependency

#4012 - Twitter Bots Example Overfits "Out-of-the-Box"

Issue - State: open - Opened by Quinn-Meyer-Sustainment 8 months ago - 1 comment
Labels: llm

#4011 - Fix mnist source

Pull Request - State: closed - Opened by mhabedank 8 months ago - 1 comment

#4010 - 4/5 trial fails due to lack of memory

Issue - State: open - Opened by diegotxegp 8 months ago - 2 comments
Labels: bug

#4009 - MNIST Dataset can't be downloaded

Issue - State: closed - Opened by mhabedank 8 months ago

#4008 - [MAINTENANCE] Refactor and clean up.

Pull Request - State: closed - Opened by alexsherstinsky 9 months ago - 1 comment

#4007 - [MAINTENANCE] Most Recent Version of matplotlib breaks ptitprince and seaborn method calls.

Pull Request - State: closed - Opened by alexsherstinsky 9 months ago - 1 comment

#4005 - Add llava support in ludwig

Issue - State: open - Opened by skanjila 9 months ago
Labels: lmm

#4001 - Docs: update visualize.py

Pull Request - State: closed - Opened by eltociear 9 months ago - 1 comment

#4000 - Error running inference on Llama3 model

Issue - State: open - Opened by vinven7 10 months ago - 5 comments
Labels: bug, llm

#3999 - phi 3 error

Issue - State: open - Opened by tahaAmineMiri 10 months ago - 7 comments
Labels: llm

#3998 - Uploading model to HF

Issue - State: closed - Opened by vinven7 10 months ago - 1 comment

#3997 - Fix Falcon padding token issue

Pull Request - State: open - Opened by arnavgarg1 10 months ago - 1 comment

#3996 - Issue fien tuning Falcon

Issue - State: open - Opened by vinven7 10 months ago - 2 comments
Labels: bug, llm

#3995 - Issues fine tuning Mistral

Issue - State: closed - Opened by vinven7 10 months ago - 1 comment

#3993 - small typo in dequantization script

Pull Request - State: closed - Opened by arnavgarg1 10 months ago - 1 comment

#3991 - Ray retraining fails with StopIteration exception when retraining a model with small datasets

Issue - State: open - Opened by vijayi1 10 months ago - 1 comment
Labels: bug, ray, dependency

#3990 - Update Development Version to Correspond to Release-0.10.3

Pull Request - State: closed - Opened by alexsherstinsky 10 months ago

#3989 - Update Ludwig version to v0.10.3.

Pull Request - State: closed - Opened by alexsherstinsky 10 months ago

#3988 - [MAINTENANCE] Replace Slack links with Discord links.

Pull Request - State: closed - Opened by alexsherstinsky 10 months ago - 1 comment

#3987 - PyYAML error while installing with python 3.12

Issue - State: open - Opened by fmatray 10 months ago - 2 comments

#3986 - Minor change to fix the incorrect response truncation

Pull Request - State: closed - Opened by amankhandelia 10 months ago - 2 comments

#3985 - `RESPONSE` contains lot longer text than is expected based on the `output_features` and `max_sequence_length`.

Issue - State: closed - Opened by amankhandelia 10 months ago - 2 comments

#3984 - Actually add support for RSLoRA and DoRA

Pull Request - State: closed - Opened by arnavgarg1 10 months ago - 1 comment

#3983 - Fix to resume adapter training from an existing adapter weights

Pull Request - State: closed - Opened by amankhandelia 10 months ago - 4 comments

#3981 - Support for freezing pretrained vision model layers with regex

Pull Request - State: closed - Opened by ethanreidel 10 months ago - 5 comments

#3980 - Dependency issue

Issue - State: closed - Opened by robhheise 10 months ago - 3 comments

#3979 - Token-level Probability Always 0.0 When Fine-tuning Llama2-7b Model on Single GPU

Issue - State: open - Opened by MoOo2mini 10 months ago - 1 comment
Labels: llm

#3978 - Wandb on ludwigai/ludwig-ray-gpu:latest + ray throws AttributeError: module 'pydantic.fields' has no attribute 'ModelField'

Issue - State: open - Opened by rahulvramesh 11 months ago
Labels: bug, llm, ray, dependency

#3977 - Fix for 'upload_to_hf_hub()' path mismatch with 'save()'

Pull Request - State: closed - Opened by sanjaydasgupta 11 months ago - 4 comments

#3976 - Pin minimum transformers to 4.39 to reduce Llama/Gemma memory pressure

Pull Request - State: closed - Opened by arnavgarg1 11 months ago - 1 comment

#3975 - update latest version for development

Pull Request - State: closed - Opened by alexsherstinsky 11 months ago

#3974 - Update ludwig version to v0.10.2

Pull Request - State: closed - Opened by alexsherstinsky 11 months ago

#3973 - <DO_NOT_MERGE>[TROUBLESHOOTING][BUGFIX]Empty test commit (in order to make pull request) so as to run tests.</DO_NOT_MERGE>

Pull Request - State: open - Opened by alexsherstinsky 11 months ago - 2 comments

#3972 - GPU is not available

Issue - State: closed - Opened by LittleStarrider 11 months ago - 2 comments

#3971 - Allow image bytes type during preprocessing

Pull Request - State: closed - Opened by vijayi1 11 months ago - 3 comments

#3970 - fix: change PROMPT constant to reduce collision

Pull Request - State: open - Opened by geoffreyangus 11 months ago - 4 comments

#3969 - enh: enable loading model weights from training checkpoint

Pull Request - State: closed - Opened by geoffreyangus 11 months ago - 3 comments

#3968 - api.py: corrected save path in 'LudwigModel.save()'

Pull Request - State: closed - Opened by sanjaydasgupta 11 months ago

#3967 - [WIP] Gradual Unfreezing to mitigate catastrophic forgetting

Pull Request - State: closed - Opened by ethanreidel 11 months ago - 2 comments

#3966 - Support for Models stored in GCS bucket

Issue - State: open - Opened by kainspraveen 11 months ago - 3 comments
Labels: waiting for answer

#3965 - Save ludwig-config with model-weights in output directory

Pull Request - State: closed - Opened by sanjaydasgupta 11 months ago - 16 comments

#3964 - Save config with weights

Pull Request - State: closed - Opened by sanjaydasgupta 11 months ago - 1 comment

#3963 - Ray - protobuf issue

Issue - State: open - Opened by robhheise 11 months ago - 5 comments
Labels: bug, ray

#3962 - Improve docker build times for `ludwig-ray` and `ludwig-ray-gpu`

Issue - State: open - Opened by arnavgarg1 11 months ago

#3961 - Add Ludwig config json to output directory containing model weights

Issue - State: closed - Opened by sanjaydasgupta 11 months ago - 3 comments

#3960 - Dependency issue

Issue - State: closed - Opened by robhheise 11 months ago - 8 comments

#3959 - [BUGFIX] Fixing integration test failures.

Pull Request - State: closed - Opened by alexsherstinsky 11 months ago - 1 comment

#3958 - enh: add batch size tuning memory limit option to hedge against OOMs

Pull Request - State: closed - Opened by geoffreyangus 11 months ago - 1 comment

#3957 - Add support for eval batch size tuning for LLMs on local backend

Pull Request - State: closed - Opened by arnavgarg1 11 months ago - 2 comments

#3956 - [MAINTENANCE] Use latest version of psutil library.

Pull Request - State: closed - Opened by alexsherstinsky 11 months ago - 1 comment

#3955 - [MAINTENANCE] Comment Out PyTorch Nightly Test

Pull Request - State: closed - Opened by alexsherstinsky 11 months ago - 1 comment

#3953 - Temporarily disable expensive text metrics

Issue - State: closed - Opened by arnavgarg1 11 months ago - 1 comment

#3951 - You are calling `save_pretrained` to a 4-bit converted model, but your `bitsandbytes` version doesn't support it.

Issue - State: open - Opened by shripadk 11 months ago - 4 comments
Labels: llm

#3950 - batch_size > 1 results in NaN loss value

Issue - State: closed - Opened by K-Mistele 11 months ago - 12 comments

#3948 - Add support for RSLoRA and DoRA

Pull Request - State: closed - Opened by arnavgarg1 11 months ago - 1 comment

#3947 - [MAINTENANCE] Update Ludwig development version.

Pull Request - State: closed - Opened by alexsherstinsky 11 months ago - 1 comment

#3946 - Update ludwig version to v0.10.1

Pull Request - State: closed - Opened by alexsherstinsky 11 months ago

#3945 - fix: use eos token in target tensor for instruction-tuning

Pull Request - State: closed - Opened by geoffreyangus 12 months ago - 1 comment

#3944 - fix: Update imdb_genre_prediction dataset yaml to match dataset

Pull Request - State: closed - Opened by jeffreyftang 12 months ago - 1 comment

#3942 - Update Ludwig development version.

Pull Request - State: closed - Opened by alexsherstinsky 12 months ago

#3941 - Ludwig release version change

Pull Request - State: closed - Opened by alexsherstinsky 12 months ago

#3940 - Pinning transformers to 4.38.1 or above in order to ensure support fo…

Pull Request - State: closed - Opened by alexsherstinsky 12 months ago - 1 comment

#3939 - Unable to fine-tune when not using quantization

Issue - State: open - Opened by simaotwx 12 months ago - 4 comments
Labels: llm

#3938 - Re-Enable AdaptionPrompt when HuggingFace releases the PEFT fix.

Issue - State: open - Opened by alexsherstinsky 12 months ago
Labels: llm

#3937 - Remove default target modules for Gemma once it's updated in PEFT

Issue - State: closed - Opened by arnavgarg1 12 months ago - 1 comment
Labels: llm

#3936 - Add default LoRA target modules for Gemma

Pull Request - State: closed - Opened by arnavgarg1 12 months ago - 1 comment

#3935 - Disabling AdaptionPrompt till PEFT is fixed.

Pull Request - State: closed - Opened by alexsherstinsky 12 months ago - 1 comment

#3934 - Update ludwig version to 0.9.4

Pull Request - State: closed - Opened by alexsherstinsky 12 months ago - 2 comments

#3932 - Retrain previously fine tuned adapter

Issue - State: open - Opened by raghavbj24 12 months ago - 4 comments
Labels: bug, llm

#3931 - Inappropriate saving of the merged fine tuned llama-2 model

Issue - State: open - Opened by raghavbj24 12 months ago - 4 comments
Labels: bug, llm

GitHub / ludwig-ai/ludwig issues and pull requests