Feature Extraction
Transformers
PyTorch
Safetensors
English
magi
Manga
Object Detection
OCR
Clustering
Diarisation
custom_code
Instructions to use ragavsachdeva/magi with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ragavsachdeva/magi with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="ragavsachdeva/magi", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("ragavsachdeva/magi", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Update modelling_magi.py
Browse files- modelling_magi.py +1 -1
modelling_magi.py
CHANGED
|
@@ -449,7 +449,7 @@ class MagiModel(PreTrainedModel):
|
|
| 449 |
affinity_matrices = []
|
| 450 |
for crop_embeddings in crop_embeddings_for_batch:
|
| 451 |
crop_embeddings = crop_embeddings / crop_embeddings.norm(dim=-1, keepdim=True)
|
| 452 |
-
affinity_matrix =
|
| 453 |
affinity_matrices.append(affinity_matrix)
|
| 454 |
return affinity_matrices
|
| 455 |
affinity_matrices = []
|
|
|
|
| 449 |
affinity_matrices = []
|
| 450 |
for crop_embeddings in crop_embeddings_for_batch:
|
| 451 |
crop_embeddings = crop_embeddings / crop_embeddings.norm(dim=-1, keepdim=True)
|
| 452 |
+
affinity_matrix = crop_embeddings @ crop_embeddings.T
|
| 453 |
affinity_matrices.append(affinity_matrix)
|
| 454 |
return affinity_matrices
|
| 455 |
affinity_matrices = []
|