Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / ludwig-ai/ludwig issues and pull requests
#3496 - int: Pin `sqlalchemy` to `1.x.x` versions
Pull Request -
State: closed - Opened by ksbrar over 1 year ago
- 1 comment
#3495 - Remove falcon from presets as it requires running untrusted code
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
- 4 comments
#3494 - Pin bitsandbytes<0.41.0
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
#3493 - Require text output feature for LLM finetuning
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 4 comments
#3492 - Require using lora adapter when performing quantized fine-tuning
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
- 1 comment
#3491 - Added CLI llama2 example
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
#3490 - Updated presets with Llama-2
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
#3489 - [fix] Multiple test fixes
Pull Request -
State: closed - Opened by jeffkinnison over 1 year ago
- 3 comments
#3488 - Raise error when using multiple features with LLMs without prompt
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 2 comments
#3487 - Added quantization parameter metadata
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
- 1 comment
#3486 - Fixed QLoRA with mutli-gpu and reduce CUDA memory pressure during eval
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
- 1 comment
#3485 - Testing new llm feature with mosaicml/mpt-7b-8k, bigscience/bloomz-3b and openlm-research/open_llama_3b_v2
Issue -
State: closed - Opened by fire over 1 year ago
- 9 comments
#3484 - Removed extra clone during prediction and added quantization error handling
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
- 1 comment
#3483 - Exclude frozen weights from checkpoints and fix evaluation using quantized LLMs
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
- 1 comment
Labels: bug
#3482 - Bug in the Tutorial of Tabular Data Classification
Issue -
State: closed - Opened by haoyuejudy over 1 year ago
- 2 comments
#3481 - Fixed divide by zero when tuning batch size
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
- 1 comment
#3480 - Add `ludwig upload` to push artifacts to HuggingFace Hub
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 1 comment
#3479 - int: Fix quantization schema
Pull Request -
State: closed - Opened by ksbrar over 1 year ago
- 1 comment
#3478 - Enable evaluation with LLMs <7B
Pull Request -
State: closed - Opened by geoffreyangus over 1 year ago
- 1 comment
#3477 - Add RoPE scaling to increase context length up to 8K for training or inference.
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 1 comment
#3476 - Add QLoRA for 4-bit fine-tuning
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
- 2 comments
#3475 - ZeroDivisionError in batch_size_tuner.py
Issue -
State: closed - Opened by xlgp2171 over 1 year ago
- 1 comment
#3474 - ValueError: Unexpected keyword arguments: `top_k`
Issue -
State: closed - Opened by guochunjiang over 1 year ago
- 2 comments
#3473 - Remove `tables` requirement as it causes issues installing ludwig in linux env.
Pull Request -
State: closed - Opened by justinxzhao over 1 year ago
- 1 comment
#3472 - Update README.md
Pull Request -
State: closed - Opened by Skizzy-create over 1 year ago
#3471 - typo in parameter of README.md
Pull Request -
State: closed - Opened by Skizzy-create over 1 year ago
#3470 - fix: Read partitioned parquet files from relative paths
Pull Request -
State: closed - Opened by jeffkinnison over 1 year ago
- 1 comment
#3469 - Zero copy initialization of models onto training workers for LLMs
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 1 comment
#3468 - Re-enable Horovod installation and unit tests for torch nightly.
Issue -
State: closed - Opened by justinxzhao over 1 year ago
#3467 - Updated Ludwig CI for torch nightly
Pull Request -
State: closed - Opened by Infernaught over 1 year ago
- 1 comment
#3466 - Unpin deepspeed
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 1 comment
#3465 - Missing documentation for Ludwig Explainer
Issue -
State: closed - Opened by Exitare over 1 year ago
- 5 comments
#3464 - [llm_text_generation] RuntimeError: Expected all tensors to be on the same device,
Issue -
State: closed - Opened by chongxiaoc over 1 year ago
- 4 comments
#3463 - Ludwig CI for torch nightly
Pull Request -
State: closed - Opened by Infernaught over 1 year ago
- 3 comments
#3462 - Image Classification: Config
Issue -
State: closed - Opened by AnaMiguelRodrigues1 over 1 year ago
- 4 comments
#3461 - [Example] Fix llm_finetune example json part
Pull Request -
State: closed - Opened by chongxiaoc over 1 year ago
- 12 comments
#3460 - Add low cpu mem support for LLM instantiation
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 1 comment
#3459 - Unpin torch nightly CI.
Pull Request -
State: closed - Opened by justinxzhao over 1 year ago
- 1 comment
#3457 - Initial implementation of DaftDataFrameEngine
Pull Request -
State: closed - Opened by jaychia over 1 year ago
- 4 comments
#3456 - pin torchmetrics to 0.11.4
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
#3455 - Adding --no-cache-dir to dockerfile pip install
Pull Request -
State: closed - Opened by noyoshi over 1 year ago
- 4 comments
#3454 - fix: Add titles for augmentation schema options
Pull Request -
State: closed - Opened by ksbrar over 1 year ago
- 1 comment
#3453 - refactor: Remove dict support for initializer fields. (2/2)
Pull Request -
State: closed - Opened by ksbrar over 1 year ago
- 1 comment
#3452 - Add param metadata to column parameter
Pull Request -
State: closed - Opened by connor-mccorm over 1 year ago
#3451 - Unpin Daft
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 1 comment
#3449 - refactor: add type hints for encoders
Pull Request -
State: closed - Opened by Dennis-Rall over 1 year ago
- 7 comments
#3448 - refactor: Remove dict support for initializer fields. (1/2)
Pull Request -
State: closed - Opened by ksbrar over 1 year ago
- 1 comment
#3447 - [llm] fix device placement issues when using CPUs and GPUs during LLM fine tuning
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 3 comments
#3446 - Not uploading confusion_matrix (and others) figure to Comet ML
Issue -
State: closed - Opened by CostaFernando over 1 year ago
- 4 comments
#3445 - upgrade ludwig-gpu docker image to pytorch 2.0.0
Pull Request -
State: closed - Opened by jppgks over 1 year ago
#3444 - feat: Disallow additional properties everywhere, enforce strict validation by default
Pull Request -
State: closed - Opened by ksbrar over 1 year ago
- 2 comments
#3443 - int: Pin `getdaft` to 0.1.6
Pull Request -
State: closed - Opened by ksbrar over 1 year ago
- 1 comment
#3442 - int: Add back `required` for input and output features to the Ludwig JSON schema
Pull Request -
State: closed - Opened by ksbrar over 1 year ago
- 2 comments
#3441 - Can't use Docker image ludwigai/ludwig-gpu. OSError: /opt/conda/lib/python3.10/site-packages/torchtext/lib/libtorchtext.so: undefined symbol: _ZN2at4_ops10select_int4callERKNS_6TensorEll
Issue -
State: closed - Opened by CostaFernando over 1 year ago
- 3 comments
#3440 - pyarrow.lib.ArrowInvalid: GetFileInfo() yielded path 'data.training.parquet/part.00000000.parquet', which is outside base dir 'data.training.parquet'
Issue -
State: closed - Opened by xxivk over 1 year ago
- 7 comments
Labels: bug, looking into it
#3439 - Disable flaky twitter bots dataset loading test.
Pull Request -
State: closed - Opened by justinxzhao over 1 year ago
- 2 comments
#3438 - Replace all non-word characters in feature names to ensure no downstream issues with external libraries.
Pull Request -
State: closed - Opened by justinxzhao over 1 year ago
- 1 comment
#3437 - [LLM] Various fixes for LLM Fine-Tuning issues that caused loss disparity between train and val sets
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 2 comments
#3436 - Update README.md, fixed some quotes
Pull Request -
State: closed - Opened by neuhausler over 1 year ago
#3435 - Add RLHF Reward Trainer and Loss
Pull Request -
State: closed - Opened by asdataminer over 1 year ago
- 1 comment
#3434 - Update UCI dataset paths
Pull Request -
State: closed - Opened by hungcs over 1 year ago
- 1 comment
#3431 - Make Reward Model Preprocessing Modifications
Pull Request -
State: closed - Opened by asdataminer over 1 year ago
- 1 comment
#3430 - Throws 'IndexError: Dimension specified as 0 but tensor has no dimensions' during training.
Issue -
State: closed - Opened by chayanray over 1 year ago
- 6 comments
#3423 - [llm] Replace `model_name` with *required* `base_model`, add preset LLM registry, update internal adapter modules
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
- 1 comment
#3417 - Set default model for semantic retrieval
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 2 comments
#3414 - fix: Count full dataset in DatasetShard.__len__
Pull Request -
State: closed - Opened by jeffkinnison over 1 year ago
- 1 comment
#3412 - Ray 2.4 Version Bump
Pull Request -
State: closed - Opened by jeffkinnison over 1 year ago
- 1 comment
#3410 - fix: Copy modules and alias weights data when using tied weights
Pull Request -
State: closed - Opened by jeffkinnison over 1 year ago
- 3 comments
#3407 - [LLM] Fine-Tuning fixes: Correct inference mode, enable weight adjustment for adapters when fine-tuning for > 1 epoch, etc.
Pull Request -
State: open - Opened by arnavgarg1 over 1 year ago
- 2 comments
#3406 - LLMFineTurningQuestions--NotToMerge
Pull Request -
State: closed - Opened by skanjila over 1 year ago
- 1 comment
#3405 - Introduce a fourth test grouping to further speed up integration tests.
Pull Request -
State: closed - Opened by justinxzhao over 1 year ago
- 1 comment
#3404 - [llm] Parameter metadata
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
- 1 comment
#3403 - [llm] Added support for bfloat16 with deepspeed
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
- 2 comments
#3402 - Llm parameter metadata
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 2 comments
#3401 - [LLM] Set additional properties to False for Adapter field
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 2 comments
#3400 - [LLM] Ensure prompt is only supported for LLM text input features
Pull Request -
State: closed - Opened by arnavgarg1 over 1 year ago
- 2 comments
#3399 - [llm] Move prompt into top-level of config
Pull Request -
State: closed - Opened by tgaddair over 1 year ago
- 3 comments
#3398 - The torchtext version 0.14.1 in the Docker image is incompatible with pytorch 2.0.0
Issue -
State: closed - Opened by abaletinskikh over 1 year ago
- 3 comments
#3397 - fix: sequence feature explain error
Pull Request -
State: closed - Opened by jeffkinnison over 1 year ago
- 1 comment
#3396 - revert: Revert "fix: Fix dataset reading for parquet directories (#3356)"
Pull Request -
State: closed - Opened by ksbrar over 1 year ago
- 1 comment
#3395 - Torch nightly CI is pinned to 2.1.0.dev20230508+cpu. Unpin.
Issue -
State: closed - Opened by justinxzhao over 1 year ago
#3394 - Daft reads PR for 0.1
Pull Request -
State: closed - Opened by jaychia over 1 year ago
- 1 comment
#3392 - Add back in support for Dictionary Class Weights
Pull Request -
State: closed - Opened by connor-mccorm over 1 year ago
- 1 comment
#3391 - Cut CI runtime down to 1 hour by splitting integration tests into 3 separate runs.
Pull Request -
State: closed - Opened by justinxzhao almost 2 years ago
- 3 comments
#3390 - Adjust dataset synthesizer timestamp min and max
Pull Request -
State: closed - Opened by abidwael almost 2 years ago
- 1 comment
#3389 - Skip broken image torchvision test with efficientnet.
Pull Request -
State: closed - Opened by justinxzhao almost 2 years ago
- 1 comment
#3388 - fix: Typo in `zscore` metadata
Pull Request -
State: closed - Opened by ksbrar almost 2 years ago
#3387 - [LLM] Add Prefix Tuning, PTuning, Lora, Adalora and Adaption Prompt fine-tuning strategies
Pull Request -
State: closed - Opened by arnavgarg1 almost 2 years ago
- 2 comments
#3386 - [LLM] Add Prefix Tuning, PTuning, LoRA, AdaLoRA and Adaption Prompt for LLM fine-tuning
Pull Request -
State: closed - Opened by tgaddair almost 2 years ago
- 2 comments
#3385 - Added LoRA tuner for HuggingFace text encoders
Pull Request -
State: closed - Opened by tgaddair almost 2 years ago
- 1 comment
#3384 - Add special handling for empty ray datasets.
Pull Request -
State: closed - Opened by justinxzhao almost 2 years ago
- 1 comment
#3383 - [LLM] Support preloaded llm models and tokenizers during LLM model instantiation
Pull Request -
State: closed - Opened by arnavgarg1 almost 2 years ago
- 1 comment
#3382 - Ensure captum's halved batch size retry fn wrapper accepts the minimum batch size of 1.
Pull Request -
State: closed - Opened by justinxzhao almost 2 years ago
- 1 comment
#3381 - Fixed DDP checkpointing to save to local disk on each node
Pull Request -
State: closed - Opened by tgaddair almost 2 years ago
- 1 comment
#3380 - Correct TorchScript Typo in Readme
Pull Request -
State: closed - Opened by samster25 almost 2 years ago
#3379 - Fix for `best.ckpt` not found
Pull Request -
State: closed - Opened by abidwael almost 2 years ago
- 2 comments
#3378 - Temporarily switch to `eval` mode when using `batch_size: 1`
Pull Request -
State: closed - Opened by abidwael almost 2 years ago
- 1 comment
#3377 - Render config CLI test fails
Issue -
State: closed - Opened by geoffreyangus almost 2 years ago
Labels: bug
#3376 - Fix MNIST datasets img path column
Pull Request -
State: closed - Opened by shawnccx almost 2 years ago
- 4 comments
#3375 - Implemented CORN loss for ordinal classification
Pull Request -
State: closed - Opened by tgaddair almost 2 years ago
- 1 comment