Text Classification
sentence-transformers
Safetensors
Transformers
English
nvembed
feature-extraction
mteb
text
text-embeddings-inference
sparse-encoder
sparse
csr
custom_code
Eval Results (legacy)
Instructions to use Y-Research-Group/CSR-NV_Embed_v2-Classification-Banking77 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use Y-Research-Group/CSR-NV_Embed_v2-Classification-Banking77 with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("Y-Research-Group/CSR-NV_Embed_v2-Classification-Banking77", trust_remote_code=True) sentences = [ "The weather is lovely today.", "It's so sunny outside!", "He drove to the stadium." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] - Transformers
How to use Y-Research-Group/CSR-NV_Embed_v2-Classification-Banking77 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="Y-Research-Group/CSR-NV_Embed_v2-Classification-Banking77", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Y-Research-Group/CSR-NV_Embed_v2-Classification-Banking77", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
datasets:
|
| 4 |
-
- mteb/
|
| 5 |
language:
|
| 6 |
- en
|
| 7 |
-
pipeline_tag: text-
|
| 8 |
library_name: sentence-transformers
|
| 9 |
tags:
|
| 10 |
- mteb
|
|
@@ -14,57 +14,33 @@ tags:
|
|
| 14 |
- sparse-encoder
|
| 15 |
- sparse
|
| 16 |
- csr
|
|
|
|
| 17 |
model-index:
|
| 18 |
-
- name:
|
| 19 |
results:
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
- type: ndcg@100
|
| 40 |
-
value: 0.81514
|
| 41 |
-
- type: ndcg@1000
|
| 42 |
-
value: 0.81692
|
| 43 |
-
- type: map@10
|
| 44 |
-
value: 0.75662
|
| 45 |
-
- type: map@100
|
| 46 |
-
value: 0.7593
|
| 47 |
-
- type: map@1000
|
| 48 |
-
value: 0.75937
|
| 49 |
-
- type: recall@10
|
| 50 |
-
value: 0.93889
|
| 51 |
-
- type: recall@100
|
| 52 |
-
value: 0.98667
|
| 53 |
-
- type: recall@1000
|
| 54 |
-
value: 1.0
|
| 55 |
-
- type: precision@1
|
| 56 |
-
value: 0.67
|
| 57 |
-
- type: precision@10
|
| 58 |
-
value: 0.106
|
| 59 |
-
- type: mrr@10
|
| 60 |
-
value: 0.76503
|
| 61 |
-
- type: main_score
|
| 62 |
-
value: 0.80426
|
| 63 |
-
task:
|
| 64 |
-
type: Retrieval
|
| 65 |
---
|
| 66 |
|
| 67 |
|
|
|
|
| 68 |
For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our [Github](https://github.com/neilwen987/CSR_Adaptive_Rep).
|
| 69 |
|
| 70 |
|
|
@@ -77,18 +53,23 @@ We recommend using ``Transformers 4.47.0.``
|
|
| 77 |
You can evaluate this model loaded by Sentence Transformers with the following code snippet:
|
| 78 |
```python
|
| 79 |
import mteb
|
| 80 |
-
from sentence_transformers import
|
| 81 |
-
model =
|
| 82 |
-
"Y-Research-Group/CSR-NV_Embed_v2-
|
| 83 |
trust_remote_code=True
|
| 84 |
)
|
| 85 |
model.prompts = {
|
| 86 |
-
"
|
| 87 |
}
|
| 88 |
-
task = mteb.get_tasks(tasks=["
|
| 89 |
evaluation = mteb.MTEB(tasks=task)
|
| 90 |
-
evaluation.run(
|
| 91 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 92 |
```
|
| 93 |
|
| 94 |
## Citation
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
datasets:
|
| 4 |
+
- mteb/banking77
|
| 5 |
language:
|
| 6 |
- en
|
| 7 |
+
pipeline_tag: text-classification
|
| 8 |
library_name: sentence-transformers
|
| 9 |
tags:
|
| 10 |
- mteb
|
|
|
|
| 14 |
- sparse-encoder
|
| 15 |
- sparse
|
| 16 |
- csr
|
| 17 |
+
|
| 18 |
model-index:
|
| 19 |
+
- name: CSR
|
| 20 |
results:
|
| 21 |
+
- dataset:
|
| 22 |
+
name: MTEB Banking77Classification
|
| 23 |
+
type: mteb/banking77
|
| 24 |
+
config: default
|
| 25 |
+
revision: 0fd18e25b25c072e09e0d92ab615fda904d66300
|
| 26 |
+
split: test
|
| 27 |
+
metrics:
|
| 28 |
+
- type: accuracy
|
| 29 |
+
value: 0.899545
|
| 30 |
+
- type: f1
|
| 31 |
+
value: 0.899018
|
| 32 |
+
- type: f1_weighted
|
| 33 |
+
value: 0.899018
|
| 34 |
+
- type: main_score
|
| 35 |
+
value: 0.899545
|
| 36 |
+
task:
|
| 37 |
+
type: Classification
|
| 38 |
+
base_model:
|
| 39 |
+
- nvidia/NV-Embed-v2
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
---
|
| 41 |
|
| 42 |
|
| 43 |
+
|
| 44 |
For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our [Github](https://github.com/neilwen987/CSR_Adaptive_Rep).
|
| 45 |
|
| 46 |
|
|
|
|
| 53 |
You can evaluate this model loaded by Sentence Transformers with the following code snippet:
|
| 54 |
```python
|
| 55 |
import mteb
|
| 56 |
+
from sentence_transformers import SparseEncoder
|
| 57 |
+
model = SparseEncoder(
|
| 58 |
+
"Y-Research-Group/CSR-NV_Embed_v2-Classification-Banking77",
|
| 59 |
trust_remote_code=True
|
| 60 |
)
|
| 61 |
model.prompts = {
|
| 62 |
+
"Banking77Classification": "Instruct: Given a online banking query, find the corresponding intents\nQuery:"
|
| 63 |
}
|
| 64 |
+
task = mteb.get_tasks(tasks=["Banking77Classification"])
|
| 65 |
evaluation = mteb.MTEB(tasks=task)
|
| 66 |
+
evaluation.run(
|
| 67 |
+
model,
|
| 68 |
+
eval_splits=["test"],
|
| 69 |
+
output_folder="./results/Banking77Classification",
|
| 70 |
+
show_progress_bar=True
|
| 71 |
+
encode_kwargs={"convert_to_sparse_tensor": False, "batch_size": 8}
|
| 72 |
+
) # MTEB don't support sparse tensors yet, so we need to convert to dense tensors
|
| 73 |
```
|
| 74 |
|
| 75 |
## Citation
|