Skip to content

Releases: huggingface/peft

Version 0.14.0: EVA, Context-aware Prompt Tuning, Bone, and more

06 Dec 11:42
@githubnemo githubnemo
v0.14.0
de88c70
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.
Compare
Choose a tag to compare
Version 0.14.0: EVA, Context-aware Prompt Tuning, Bone, and more Latest
Latest

Highlights

peft-v0 14 0

New Methods

Context-aware Prompt Tuning

@tsachiblau added a new soft prompt method called Context-aware Prompt Tuning (CPT) which is a combination of In-Context Learning and Prompt Tuning in the sense that, for each training sample, it builds a learnable context from training examples in addition to the single training sample. Allows for sample- and parameter-efficient few-shot classification and addresses recency-bias.

Explained Variance Adaptation

@sirluk contributed a new LoRA initialization method called Explained Variance Adaptation (EVA). Instead of randomly initializing LoRA weights, this method uses SVD of the base layer weights to initialize the LoRA weights and is also able to re-allocate the ranks of the adapter based on the explained variance ratio (derived from SVD). Thus, this initialization method can yield better initial values and better rank distribution.

Bone

@JL-er added an implementation for Block Affine (Bone) Adaptation which utilizes presumed sparsity in the base layer weights to divide them into multiple sub-spaces that share a single low-rank matrix for updates. Compared to LoRA, Bone has the potential to significantly reduce memory usage and achieve faster computation.

Enhancements

PEFT now supports LoRAs for int8 torchao quantized models (check this and this notebook) . In addition, VeRA can now be used with 4 and 8 bit bitsandbytes quantization thanks to @ZiadHelal.

Hot-swapping of LoRA adapters is now possible using the hotswap_adapter function. Now you are able to load one LoRA and replace its weights in-place with the LoRA weights of another adapter which, in general, should be faster than deleting one adapter and loading the other adapter in its place. The feature is built so that no re-compilation of the model is necessary if torch.compile was called on the model (right now, this requires ranks and alphas to be the same for the adapters).

LoRA and IA³ now support Conv3d layers thanks to @jsilter, and @JINO-ROHIT added a notebook showcasing PEFT model evaluation using lm-eval-harness toolkit.

With the target_modules argument, you can specify which layers to target with the adapter (e.g. LoRA). Now you can also specify which modules not to target by using the exclude_modules parameter (thanks @JINO-ROHIT).

Changes

  • There have been made several fixes to the OFT implementation, among other things, to fix merging, which makes adapter weights trained with PEFT versions prior to this release incompatible (see #1996 for details).
  • Adapter configs are now forward-compatible by accepting unknown keys.
  • Prefix tuning was fitted to the DynamicCache caching infrastructure of transformers (see #2096). If you are using this PEFT version and a recent version of transformers with an old prefix tuning checkpoint, you should double check that it still works correctly and retrain it if it doesn't.
  • Added lora_bias parameter to LoRA layers to enable bias on LoRA B matrix. This is useful when extracting LoRA weights from fully fine-tuned parameters with bias vectors so that these can be taken into account.
  • #2180 provided a couple of bug fixes to LoKr (thanks @yaswanth19). If you're using LoKr, your old checkpoints should still work but it's recommended to retrain your adapter.
  • from_pretrained now warns the user if PEFT keys are missing.
  • Attribute access to modules in modules_to_save is now properly and transparently handled.
  • PEFT supports the changes to bitsandbytes 8bit quantization from the recent v0.45.0 release. To benefit from these improvements, we thus recommend to upgrade bitsandbytes if you're using QLoRA. Expect slight numerical differences in model outputs if you're using QLoRA with 8bit bitsandbytes quantization.

What's Changed

Read more

Contributors

githubnemo, jsilter, and 19 other contributors
Assets 2
Loading

v0.13.2: Small patch release

11 Oct 11:45
@BenjaminBossan BenjaminBossan
v0.13.2
431c0e2
Compare
Choose a tag to compare
v0.13.2: Small patch release

This patch release contains a small bug fix for an issue that prevented some LoRA checkpoints to be loaded correctly (mostly concerning stable diffusion checkpoints not trained with PEFT when loaded in diffusers, #2144).

Full Changelog: v0.13.1...v0.13.2

Assets 2
Loading

v0.13.1: Small patch release

08 Oct 12:29
@BenjaminBossan BenjaminBossan
v0.13.1
b8da272
Compare
Choose a tag to compare
v0.13.1: Small patch release

This patch release contains a small bug fix for the low_cpu_mem_usage=True option (#2113).

Full Changelog: v0.13.0...v0.13.1

Assets 2
Loading

v0.13.0: LoRA+, VB-LoRA, and more

25 Sep 12:11
@BenjaminBossan BenjaminBossan
v0.13.0
f0b066e
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.
Compare
Choose a tag to compare
v0.13.0: LoRA+, VB-LoRA, and more

peft-v0 13 0

Highlights

New methods

LoRA+

@kallewoof added LoRA+ to PEFT (#1915). This is a function that allows to initialize an optimizer with settings that are better suited for training a LoRA adapter.

VB-LoRA

@leo-yangli added a new method to PEFT called VB-LoRA (#2039). The idea is to have LoRA layers be composed from a single vector bank (hence "VB") that is shared among all layers. This makes VB-LoRA extremely parameter efficient and the checkpoints especially small (comparable to the VeRA method), while still promising good fine-tuning performance. Check the VB-LoRA docs and example.

Enhancements

New Hugging Face team member @ariG23498 added the helper function rescale_adapter_scale to PEFT (#1951). Use this context manager to temporarily increase or decrease the scaling of the LoRA adapter of a model. It also works for PEFT adapters loaded directly into a transformers or diffusers model.

@ariG23498 also added DoRA support for embedding layers (#2006). So if you're using the use_dora=True option in the LoraConfig, you can now also target embedding layers.

For some time now, we support inference with batches that are using different adapters for different samples, so e.g. sample 1-5 use "adapter1" and samples 6-10 use "adapter2". However, this only worked for LoRA layers so far. @saeid93 extended this to also work with layers targeted by modules_to_save (#1990).

When loading a PEFT adapter, you now have the option to pass low_cpu_mem_usage=True (#1961). This will initialize the adapter with empty weights ("meta" device) before loading the weights instead of initializing on CPU or GPU. This can speed up loading PEFT adapters. So use this option especially if you have a lot of adapters to load at the same time or if these adapters are very big. Please let us know if you encounter issues with this option, as we may make this the default in the future.

Changes

Safe loading of PyTorch weights

Unless indicated otherwise, PEFT adapters are saved and loaded using the secure safetensors format. However, we also support the PyTorch format for checkpoints, which relies on the inherently insecure pickle protocol from Python. In the future, PyTorch will be more strict when loading these files to improve security by making the option weights_only=True the default. This is generally recommended and should not cause any trouble with PEFT checkpoints, which is why with this release, PEFT will enable this by default. Please open an issue if this causes trouble.

What's Changed

Read more

Contributors

kallewoof, keakon, and 20 other contributors
Assets 2
Loading

v0.12.0: New methods OLoRA, X-LoRA, FourierFT, HRA, and much more

24 Jul 11:55
@BenjaminBossan BenjaminBossan
v0.12.0
e6cd24c
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.
Compare
Choose a tag to compare
v0.12.0: New methods OLoRA, X-LoRA, FourierFT, HRA, and much more

Highlights

peft-v0 12 0

New methods

OLoRA

@tokenizer-decode added support for a new LoRA initialization strategy called OLoRA (#1828). With this initialization option, the LoRA weights are initialized to be orthonormal, which promises to improve training convergence. Similar to PiSSA, this can also be applied to models quantized with bitsandbytes. Check out the accompanying OLoRA examples.

X-LoRA

@EricLBuehler added the X-LoRA method to PEFT (#1491). This is a mixture of experts approach that combines the strength of multiple pre-trained LoRA adapters. Documentation has yet to be added but check out the X-LoRA tests for how to use it.

FourierFT

@Phoveran, @zqgao22, @Chaos96, and @DSAILatHKUST added discrete Fourier transform fine-tuning to PEFT (#1838). This method promises to match LoRA in terms of performance while reducing the number of parameters even further. Check out the included FourierFT notebook.

HRA

@DaShenZi721 added support for Householder Reflection Adaptation (#1864). This method bridges the gap between low rank adapters like LoRA on the one hand and orthogonal fine-tuning techniques such as OFT and BOFT on the other. As such, it is interesting for both LLMs and image generation models. Check out the HRA example on how to perform DreamBooth fine-tuning.

Enhancements

  • IA³ now supports merging of multiple adapters via the add_weighted_adapter method thanks to @alexrs (#1701).
  • Call peft_model.get_layer_status() and peft_model.get_model_status() to get an overview of the layer/model status of the PEFT model. This can be especially helpful when dealing with multiple adapters or for debugging purposes. More information can be found in the docs (#1743).
  • DoRA now supports FSDP training, including with bitsandbytes quantization, aka QDoRA ()#1806).
  • VeRA has been extended by @dkopi to support targeting layers with different weight shapes (#1817).
  • @kallewoof added the possibility for ephemeral GPU offloading. For now, this is only implemented for loading DoRA models, which can be sped up considerably for big models at the cost of a bit of extra VRAM (#1857).
  • Experimental: It is now possible to tell PEFT to use your custom LoRA layers through dynamic dispatching. Use this, for instance, to add LoRA layers for thus far unsupported layer types without the need to first create a PR on PEFT (but contributions are still welcome!) (#1875).

Examples

Changes

Casting of the adapter dtype

Important: If the base model is loaded in float16 (fp16) or bfloat16 (bf16), PEFT now autocasts adapter weights to float32 (fp32) instead of using the dtype of the base model (#1706). This requires more memory than previously but stabilizes training, so it's the more sensible default. To prevent this, pass autocast_adapter_dtype=False when calling get_peft_model, PeftModel.from_pretrained, or PeftModel.load_adapter.

Adapter device placement

The logic of device placement when loading multiple adapters on the same model has been changed (#1742). Previously, PEFT would move all adapters to the device of the base model. Now, only the newly loaded/created adapter is moved to the base model's device. This allows users to have more fine-grained control over the adapter devices, e.g. allowing them to offload unused adapters to CPU more easily.

PiSSA

  • Calling save_pretrained with the convert_pissa_to_lora argument is deprecated, the argument was renamed to path_initial_model_for_weight_conversion (#1828). Also, calling this no longer deletes the original adapter (#1933).
  • Using weight conversion (path_initial_model_for_weight_conversion) while also using use_rslora=True and rank_pattern or alpha_pattern now raises an error (#1930). This used not to raise but inference would return incorrect outputs. We also warn about this setting during initialization.

Call for contributions

We are now making sure to tag appropriate issues with the contributions welcome label. If you are looking for a way to contribute to PEFT, check out these issues.

What's Changed

Read more

Contributors

kallewoof, alexrs, and 30 other contributors
Assets 2
Loading

v0.11.1

17 May 12:55
@BenjaminBossan BenjaminBossan
v0.11.1
207376d
Compare
Choose a tag to compare
v0.11.1

Patch release v0.11.1

Fix a bug that could lead to C++ compilation errors after importing PEFT (#1738 #1739).

Full Changelog: v0.11.0...v0.11.1

Assets 2
Loading

v0.11.0: New PEFT methods BOFT, VeRA, PiSSA, quantization with HQQ and EETQ, and more

16 May 09:53
@BenjaminBossan BenjaminBossan
v0.11.0
0649947
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.
Compare
Choose a tag to compare
v0.11.0: New PEFT methods BOFT, VeRA, PiSSA, quantization with HQQ and EETQ, and more

Highlights

peft-v0 11 0

New methods

BOFT

Thanks to @yfeng95, @Zeju1997, and @YuliangXiu, PEFT was extended with BOFT: Parameter-Efficient Orthogonal Finetuning via Butterfly Factorization (#1326, BOFT paper link). In PEFT v0.7.0, we already added OFT, but BOFT is even more parameter efficient. Check out the included BOFT controlnet and BOFT dreambooth examples.

VeRA

If the parameter reduction of LoRA is not enough for your use case, you should take a close look at VeRA: Vector-based Random Matrix Adaptation (#1564, VeRA paper link). This method resembles LoRA but adds two learnable scaling vectors to the two LoRA weight matrices. However, the LoRA weights themselves are shared across all layers, considerably reducing the number of trainable parameters.

The bulk of this PR was implemented by contributor @vvvm23 with the help of @dkopi.

PiSSA

PiSSA, Principal Singular values and Singular vectors Adaptation, is a new initialization method for LoRA, which was added by @fxmeng (#1626, PiSSA paper link). The improved initialization promises to speed up convergence and improve the final performance of LoRA models. When using models quantized with bitsandbytes, PiSSA initialization should reduce the quantization error, similar to LoftQ.

Quantization

HQQ

Thanks to @fahadh4ilyas, PEFT LoRA linear layers now support Half-Quadratic Quantization, HQQ (#1618, HQQ repo). HQQ is fast and efficient (down to 2 bits), while not requiring calibration data.

EETQ

Another new quantization method supported in PEFT is Easy & Efficient Quantization for Transformers, EETQ (#1675, EETQ repo). This 8 bit quantization method works for LoRA linear layers and should be faster than bitsandbytes.

Show adapter layer and model status

We added a feature to show adapter layer and model status of PEFT models in #1663. With the newly added methods, you can easily check what adapters exist on your model, whether gradients are active, whether they are enabled, which ones are active or merged. You will also be informed if irregularities have been detected.

To use this new feature, call model.get_layer_status() for layer-level information, and model.get_model_status() for model-level information. For more details, check out our docs on layer and model status.

Changes

Edge case of how we deal with modules_to_save

We had the issue that when we were using classes such as PeftModelForSequenceClassification, we implicitly added the classifier layers to model.modules_to_save. However, this would only add a new ModulesToSaveWrapper instance for the first adapter being initialized. When initializing a 2nd adapter via model.add_adapter, this information was ignored. Now, peft_config.modules_to_save is updated explicitly to add the classifier layers (#1615). This is a departure from how this worked previously, but it reflects the intended behavior better.

Furthermore, when merging together multiple LoRA adapters using model.add_weighted_adapter, if these adapters had modules_to_save, the original parameters of these modules would be used. This is unexpected and will most likely result in bad outputs. As there is no clear way to merge these modules, we decided to raise an error in this case (#1615).

What's Changed

Read more

Contributors

winglian, charliermarsh, and 24 other contributors
Assets 2
Loading

v0.10.0: Fine-tune larger QLoRA models with DeepSpeed and FSDP, layer replication, enhance DoRA

21 Mar 10:20
@BenjaminBossan BenjaminBossan
v0.10.0
8221246
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.
Compare
Choose a tag to compare
v0.10.0: Fine-tune larger QLoRA models with DeepSpeed and FSDP, layer replication, enhance DoRA

Highlights

image

Support for QLoRA with DeepSpeed ZeRO3 and FSDP

We added a couple of changes to allow QLoRA to work with DeepSpeed ZeRO3 and Fully Sharded Data Parallel (FSDP). For instance, this allows you to fine-tune a 70B Llama model on two GPUs with 24GB memory each. Besides the latest version of PEFT, this requires bitsandbytes>=0.43.0, accelerate>=0.28.0, transformers>4.38.2, trl>0.7.11. Check out our docs on DeepSpeed and FSDP with PEFT, as well as this blogpost from answer.ai, for more details.

Layer replication

First time contributor @siddartha-RE added support for layer replication with LoRA. This allows you to duplicate layers of a model and apply LoRA adapters to them. Since the base weights are shared, this costs only very little extra memory, but can lead to a nice improvement of model performance. Find out more in our docs.

Improving DoRA

Last release, we added the option to enable DoRA in PEFT by simply adding use_dora=True to your LoraConfig. However, this only worked for non-quantized linear layers. With this PEFT release, we now also support Conv2d layers, as well as linear layers quantized with bitsandbytes.

Mixed LoRA adapter batches

If you have a PEFT model with multiple LoRA adapters attached to it, it's now possible to apply different adapters (or, in fact, no adapter) on different samples in the same batch. To do this, pass a list of adapter names as an additional argument. For example, if you have a batch of three samples:

output = model(**inputs, adapter_names=["adapter1", "adapter2", "__base__"])`

Here, "adapter1" and "adapter2" should be the same name as your corresponding LoRA adapters and "__base__" is a special name that refers to the base model without any adapter. Find more details in our docs.

Without this feature, if you wanted to run inference with different LoRA adapters, you'd have to use single samples or try to group batches with the same adapter, then switch between adapters using set_adapter -- this is inefficient and inconvenient. Therefore, it is recommended to use this new, faster method from now on when encountering this scenario.

New LoftQ initialization function

We added an alternative way to initialize LoRA weights for a quantized model using the LoftQ method, which can be more convenient than the existing method. Right now, using LoftQ requires you to go through multiple steps as shown here. Furthermore, it's necessary to keep a separate copy of the quantized weights, as those are not identical to the quantized weights from the default model.

Using the new replace_lora_weights_loftq function, it's now possible to apply LoftQ initialization in a single step and without the need for extra copies of the weights. Check out the docs and this example notebook to see how it works. Right now, this method only supports 4bit quantization with bitsandbytes, and the model has to be stored in the safetensors format.

Deprecations

The function prepare_model_for_int8_training was deprecated for quite some time and is now removed completely. Use prepare_model_for_kbit_training instead.

What's Changed

Besides these highlights, we added many small improvements and fixed a couple of bugs. All these changes are listed below. As always, we thank all the awesome contributors who helped us improve PEFT.

New Contributors

Full Changelog: v0.9.0...v0.10.0

Contributors

BenjaminBossan, pacman100, and 9 other contributors
Assets 2
Loading

v0.9.0: Merging LoRA weights, new quantization options, DoRA support, and more

28 Feb 10:37
@BenjaminBossan BenjaminBossan
v0.9.0
7e5335d
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.
Compare
Choose a tag to compare
v0.9.0: Merging LoRA weights, new quantization options, DoRA support, and more

Highlights

New methods for merging LoRA weights together

cat_teapot

With PR #1364, we added new methods for merging LoRA weights together. This is not about merging LoRA weights into the base model. Instead, this is about merging the weights from different LoRA adapters into a single adapter by calling add_weighted_adapter. This allows you to combine the strength from multiple LoRA adapters into a single adapter, while being faster than activating each of these adapters individually.

Although this feature has already existed in PEFT for some time, we have added new merging methods that promise much better results. The first is based on TIES, the second on DARE and a new one inspired by both called Magnitude Prune. If you haven't tried these new methods, or haven't touched the LoRA weight merging feature at all, you can find more information here:

AWQ and AQLM support for LoRA

Via #1394, we now support AutoAWQ in PEFT. This is a new method for 4bit quantization of model weights.

Screenshot 2024-02-28 at 09 41 40

Similarly, we now support AQLM via #1476. This method allows to quantize weights to as low as 2 bits. Both methods support quantizing nn.Linear layers. To find out more about all the quantization options that work with PEFT, check out our docs here.

Screenshot 2024-02-28 at 09 42 22

Note these integrations do not support merge_and_unload() yet, meaning for inference you need to always attach the adapter weights into the base model

DoRA support

We now support Weight-Decomposed Low-Rank Adaptation aka DoRA via #1474. This new method is builds on top of LoRA and has shown very promising results. Especially at lower ranks (e.g. r=8), it should perform much better than LoRA. Right now, only non-quantized nn.Linear layers are supported. If you'd like to give it a try, just pass use_dora=True to your LoraConfig and you're good to go.

Documentation

Thanks to @stevhliu and many other contributors, there have been big improvements to the documentation. You should find it more organized and more up-to-date. Our DeepSpeed and FSDP guides have also been much improved.

Check out our improved docs if you haven't already!

Development

If you're implementing custom adapter layers, for instance a custom LoraLayer, note that all subclasses should now implement update_layer -- unless they want to use the default method by the parent class. In particular, this means you should no longer use different method names for the subclass, like update_layer_embedding. Also, we generally don't permit ranks (r) of 0 anymore. For more, see this PR.

Developers should have an easier time now since we fully embrace ruff. If you're the type of person who forgets to call make style before pushing to a PR, consider adding a pre-commit hook. Tests are now a bit less verbose by using plain asserts and generally embracing pytest features more fully. All of this comes thanks to @akx.

What's Changed

On top of these changes, we have added a lot of small changes since the last release, check out the full changes below. As always, we had a lot of support by many contributors, you're awesome!

Read more

Contributors

akx, szepeviktor, and 14 other contributors
Assets 2
Loading

Release v0.8.2

01 Feb 14:16
@pacman100 pacman100
v0.8.2
e37bff6
Compare
Choose a tag to compare
Release v0.8.2

What's Changed

New Contributors

Full Changelog: v0.8.1...v0.8.2

Contributors

BenjaminBossan, pacman100, and 5 other contributors
Assets 2
Loading