Instructions to use Lin-Chen/ShareCaptioner with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Lin-Chen/ShareCaptioner with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="Lin-Chen/ShareCaptioner", trust_remote_code=True, device_map="auto")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Lin-Chen/ShareCaptioner", trust_remote_code=True, dtype="auto", device_map="auto") - Notebooks
- Google Colab
- Kaggle
chenlin commited on
Commit ·
8f9c756
1
Parent(s): a285f80
fix bug
Browse files- resampler.py +1 -1
resampler.py
CHANGED
|
@@ -44,7 +44,7 @@ def get_1d_sincos_pos_embed_from_grid(embed_dim, pos):
|
|
| 44 |
out: (M, D)
|
| 45 |
"""
|
| 46 |
assert embed_dim % 2 == 0
|
| 47 |
-
omega = np.arange(embed_dim // 2, dtype=np.
|
| 48 |
omega /= embed_dim / 2.
|
| 49 |
omega = 1. / 10000**omega # (D/2,)
|
| 50 |
|
|
|
|
| 44 |
out: (M, D)
|
| 45 |
"""
|
| 46 |
assert embed_dim % 2 == 0
|
| 47 |
+
omega = np.arange(embed_dim // 2, dtype=np.float32)
|
| 48 |
omega /= embed_dim / 2.
|
| 49 |
omega = 1. / 10000**omega # (D/2,)
|
| 50 |
|