Instructions to use Marqo/marqo-gcl-e5-large-v2-130 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Marqo/marqo-gcl-e5-large-v2-130 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="Marqo/marqo-gcl-e5-large-v2-130")# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("Marqo/marqo-gcl-e5-large-v2-130") model = AutoModel.from_pretrained("Marqo/marqo-gcl-e5-large-v2-130") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,6 +1,24 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
```python
|
| 5 |
import torch.nn.functional as F
|
| 6 |
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
### Generalized Contrastive Learning for Multi-Modal Retrieval and Ranking
|
| 6 |
+
|
| 7 |
+
This work aims to improve and measure the ranking performance of information retrieval models, especially for retrieving relevant products given a search query.
|
| 8 |
+
|
| 9 |
+
Blog post: https://www.marqo.ai/blog/generalized-contrastive-learning-for-multi-modal-retrieval-and-ranking
|
| 10 |
+
|
| 11 |
+
Paper: https://arxiv.org/pdf/2404.08535.pdf
|
| 12 |
+
|
| 13 |
+
### Text-only
|
| 14 |
+
| Methods | Models | nDCG | ERR | RBP |
|
| 15 |
+
|---------------|----------------------------|-----------|------------|-----------|
|
| 16 |
+
| BM25 | - | 0.071 | 0.028 | 0.052 |
|
| 17 |
+
| E5 | e5-large-v2 | 0.335 | 0.095 | 0.289 |
|
| 18 |
+
| E5 (GCL) | e5-large-v2 | **0.470** | **0.457** | **0.374** |
|
| 19 |
+
|
| 20 |
+
### Usage
|
| 21 |
+
|
| 22 |
```python
|
| 23 |
import torch.nn.functional as F
|
| 24 |
|