File size: 1,219 Bytes
22a9232 1f874ab 22a9232 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 |
Graphcore and Hugging Face are working together to make training of Transformer models on IPUs fast and easy. Learn more about how to take advantage of the power of Graphcore IPUs to train Transformers models at [hf.co/hardware/graphcore](https://huggingface.co/hardware/graphcore).
# HuBERT-Base model IPU config
This model contains just the `IPUConfig` files for running the HuBERT-base model (e.g. [facebook/hubert-base-ls960](https://huggingface.co/facebook/hubert-base-ls960)) on Graphcore IPUs.
**This model contains no model weights, only an IPUConfig.**
## Model description
HUBERT (Hidden-Unit BERT) is a BERT-based model for self-supervised speech representation learning approach that relies on predicting K-means cluster assignments of masked segments of continuous output. This approach forces the model to learn a combined acoustic and language model over the continuous inputs by applying the prediction loss over the masked region.
Paper link : [Self-Supervised Speech Representation
Learning by Masked Prediction of Hidden Unit](https://arxiv.org/pdf/2106.07447v1.pdf)
## Usage
```
from optimum.graphcore import IPUConfig
ipu_config = IPUConfig.from_pretrained("Graphcore/hubert-base-ipu")
``` |