Instructions to use RationAI/LSP-DETR with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use RationAI/LSP-DETR with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-segmentation", model="RationAI/LSP-DETR", trust_remote_code=True)# Load model directly from transformers import AutoModelForObjectDetection model = AutoModelForObjectDetection.from_pretrained("RationAI/LSP-DETR", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Upload model
Browse files- modeling.py +0 -1
modeling.py
CHANGED
|
@@ -106,7 +106,6 @@ class CayleySTRING(nn.Module):
|
|
| 106 |
positions ([b, n, pos_dim]): Positions tensor.
|
| 107 |
"""
|
| 108 |
# Compute (I + S)^-1 @ x
|
| 109 |
-
print(self.training)
|
| 110 |
if self.training:
|
| 111 |
# Use linalg.solve during training for numerical stability.
|
| 112 |
y = torch.linalg.solve(
|
|
|
|
| 106 |
positions ([b, n, pos_dim]): Positions tensor.
|
| 107 |
"""
|
| 108 |
# Compute (I + S)^-1 @ x
|
|
|
|
| 109 |
if self.training:
|
| 110 |
# Use linalg.solve during training for numerical stability.
|
| 111 |
y = torch.linalg.solve(
|