add load instructions
Browse files
README.md
CHANGED
|
@@ -15,6 +15,22 @@ Done as part of Huggingface community training ([see forum post](https://discuss
|
|
| 15 |
|
| 16 |
Builds on T5, using an autoencoder to convert it into an MMD-VAE.
|
| 17 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
## Setup
|
| 19 |
|
| 20 |
-
Run `setup_tpu_vm_venv.sh` to setup a virtual enviroment.
|
|
|
|
| 15 |
|
| 16 |
Builds on T5, using an autoencoder to convert it into an MMD-VAE.
|
| 17 |
|
| 18 |
+
## How to use from the 🤗/transformers library
|
| 19 |
+
|
| 20 |
+
Clone the model repo.
|
| 21 |
+
```bash
|
| 22 |
+
git clone https://github.com/Fraser-Greenlee/t5-vae-flax.git
|
| 23 |
+
```
|
| 24 |
+
|
| 25 |
+
```python
|
| 26 |
+
from transformers import AutoTokenizer
|
| 27 |
+
from t5_vae_flax.src.t5_vae import FlaxT5VaeForAutoencoding
|
| 28 |
+
|
| 29 |
+
tokenizer = AutoTokenizer.from_pretrained("t5-base")
|
| 30 |
+
|
| 31 |
+
model = T5VaeForAutoencoding.from_pretrained("flax-community/t5-vae-python")
|
| 32 |
+
```
|
| 33 |
+
|
| 34 |
## Setup
|
| 35 |
|
| 36 |
+
Run `setup_tpu_vm_venv.sh` to setup a virtual enviroment on a TPU VM for training.
|