π€ Optimum Habana is the interface between the π€ Transformers and Diffusers libraries and Habana's Gaudi processor (HPU). It provides a set of tools enabling easy model loading, training and inference on single- and multi-HPU settings for different downstream tasks. The list of officially validated models and tasks is available here. Users can try other models and tasks with only few changes.
HPUs offer fast model training and inference as well as a great price-performance ratio. Check out this blog post about BERT pre-training and this article benchmarking Habana Gaudi2 versus Nvidia A100 GPUs for concrete examples. If you are not familiar with HPUs and would like to know more about them, we recommend you take a look at our conceptual guide.
To install the latest stable release of this package
pip install --upgrade-strategy eager optimum[habana]
The --upgrade-strategy eager
option is needed to ensure optimum-habana
is upgraded to the latest stable release.
To use the example associated with the latest stable release, run:
git clone https://github.com/huggingface/optimum-habana cd optimum-habana && git checkout v1.10.4
with
v1.10.4
the version number of this release.
Optimum Habana is a fast-moving project, and you may want to install it from source and get the latest scripts :
pip install git+https://github.com/huggingface/optimum-habana.git
git clone https://github.com/huggingface/optimum-habana
To use DeepSpeed on HPUs, you also need to run the following command:
pip install git+https://github.com/HabanaAI/[email protected]
To install the requirements for every example:
cd <example-folder> pip install -r requirements.txt
π€ Optimum Habana was designed with one goal in mind: to make training and inference straightforward for any π€ Transformers and π€ Diffusers user while leveraging the complete power of Gaudi processors.
There are two main classes one needs to know:
- GaudiTrainer: the trainer class that takes care of compiling and distributing the model to run on HPUs, and performing training and evaluation.
- GaudiConfig: the class that enables to configure Habana Mixed Precision and to decide whether optimized operators and optimizers should be used or not.
The GaudiTrainer is very similar to the π€ Transformers Trainer, and adapting a script using the Trainer to make it work with Gaudi will mostly consist in simply swapping the Trainer
class for the GaudiTrainer
one.
That's how most of the example scripts were adapted from their original counterparts.
Here is an example:
- from transformers import Trainer, TrainingArguments
+ from optimum.habana import GaudiConfig, GaudiTrainer, GaudiTrainingArguments
- training_args = TrainingArguments(
+ training_args = GaudiTrainingArguments(
# training arguments...
+ use_habana=True,
+ use_lazy_mode=True, # whether to use lazy or eager mode
+ gaudi_config_name=path_to_gaudi_config,
)
# A lot of code here
# Initialize our Trainer
- trainer = Trainer(
+ trainer = GaudiTrainer(
model=model,
args=training_args, # Original training arguments.
train_dataset=train_dataset if training_args.do_train else None,
eval_dataset=eval_dataset if training_args.do_eval else None,
compute_metrics=compute_metrics,
tokenizer=tokenizer,
data_collator=data_collator,
)
where gaudi_config_name
is the name of a model from the Hub (Gaudi configurations are stored in model repositories) or a path to a local Gaudi configuration file (you can see here how to write your own).
You can generate images from prompts using Stable Diffusion on Gaudi using the GaudiStableDiffusionPipeline
class and the [GaudiDDIMScheduler
] which have been both optimized for HPUs. Here is how to use them and the differences with the π€ Diffusers library:
- from diffusers import DDIMScheduler, StableDiffusionPipeline
+ from optimum.habana.diffusers import GaudiDDIMScheduler, GaudiStableDiffusionPipeline
model_name = "runwayml/stable-diffusion-v1-5"
- scheduler = DDIMScheduler.from_pretrained(model_name, subfolder="scheduler")
+ scheduler = GaudiDDIMScheduler.from_pretrained(model_name, subfolder="scheduler")
- pipeline = StableDiffusionPipeline.from_pretrained(
+ pipeline = GaudiStableDiffusionPipeline.from_pretrained(
model_name,
scheduler=scheduler,
+ use_habana=True,
+ use_hpu_graphs=True,
+ gaudi_config="Habana/stable-diffusion",
)
outputs = generator(
["An image of a squirrel in Picasso style"],
num_images_per_prompt=16,
+ batch_size=4,
)
Check out the documentation of Optimum Habana for more advanced usage.
The following model architectures, tasks and device distributions have been validated for π€ Optimum Habana:
In the tables below, βοΈ means single-card, multi-card and DeepSpeed have all been validated.
- Transformers:
- Diffusers:
Architecture | Training | Inference | Tasks |
---|---|---|---|
Stable Diffusion | |||
LDM3D |
- TRL:
Architecture | Training | Inference | Tasks |
---|---|---|---|
Llama 2 | βοΈ | ||
Llama 2 | βοΈ |
Other models and tasks supported by the π€ Transformers and π€ Diffusers library may also work. You can refer to this section for using them with π€ Optimum Habana. Besides, this page explains how to modify any example from the π€ Transformers library to make it work with π€ Optimum Habana.
If you find any issues while using those, please open an issue or a pull request.
Please refer to Habana Gaudi's official installation guide.
Tests should be run in a Docker container based on Habana Docker images.
The current version has been validated for SynapseAI 1.14.
Check the contributor guide for instructions.