Instructions to use Respeecher/ukrainian-data2vec with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Respeecher/ukrainian-data2vec with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="Respeecher/ukrainian-data2vec")# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("Respeecher/ukrainian-data2vec") model = AutoModel.from_pretrained("Respeecher/ukrainian-data2vec") - Notebooks
- Google Colab
- Kaggle
Commit ·
bcc1b9a
1
Parent(s): 2cc9475
Upload model
Browse files- config.json +2 -1
config.json
CHANGED
|
@@ -1,4 +1,5 @@
|
|
| 1 |
{
|
|
|
|
| 2 |
"activation_dropout": 0.1,
|
| 3 |
"adapter_kernel_size": 3,
|
| 4 |
"adapter_stride": 2,
|
|
@@ -102,7 +103,7 @@
|
|
| 102 |
1
|
| 103 |
],
|
| 104 |
"torch_dtype": "float32",
|
| 105 |
-
"transformers_version": "4.
|
| 106 |
"use_weighted_layer_sum": false,
|
| 107 |
"vocab_size": 32,
|
| 108 |
"xvector_output_dim": 512
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "/home/vova/src/simplecatalystflow_accent_vc_v2/feature_models/data2vec-audio-large-ukrainian/",
|
| 3 |
"activation_dropout": 0.1,
|
| 4 |
"adapter_kernel_size": 3,
|
| 5 |
"adapter_stride": 2,
|
|
|
|
| 103 |
1
|
| 104 |
],
|
| 105 |
"torch_dtype": "float32",
|
| 106 |
+
"transformers_version": "4.27.0",
|
| 107 |
"use_weighted_layer_sum": false,
|
| 108 |
"vocab_size": 32,
|
| 109 |
"xvector_output_dim": 512
|