Text Ranking
sentence-transformers
Safetensors
deberta-v2
cross-encoder
reranker
Generated from Trainer
dataset_size:7419
loss:BinaryCrossEntropyLoss
Eval Results (legacy)
text-embeddings-inference
Instructions to use ColeH0415/comp90042-crossencoder-factcheck with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use ColeH0415/comp90042-crossencoder-factcheck with sentence-transformers:
from sentence_transformers import CrossEncoder model = CrossEncoder("ColeH0415/comp90042-crossencoder-factcheck") query = "Which planet is known as the Red Planet?" passages = [ "Venus is often called Earth's twin because of its similar size and proximity.", "Mars, known for its reddish appearance, is often referred to as the Red Planet.", "Jupiter, the largest planet in our solar system, has a prominent red spot.", "Saturn, famous for its rings, is sometimes mistaken for the Red Planet." ] scores = model.predict([(query, passage) for passage in passages]) print(scores) - Notebooks
- Google Colab
- Kaggle
CE fine-tuned epoch 1/3 best_val=0.8170
Browse files- README.md +39 -39
- model.safetensors +1 -1
README.md
CHANGED
|
@@ -4,7 +4,7 @@ tags:
|
|
| 4 |
- cross-encoder
|
| 5 |
- reranker
|
| 6 |
- generated_from_trainer
|
| 7 |
-
- dataset_size:
|
| 8 |
- loss:BinaryCrossEntropyLoss
|
| 9 |
base_model: cross-encoder/ms-marco-MiniLM-L6-v2
|
| 10 |
pipeline_tag: text-ranking
|
|
@@ -28,25 +28,25 @@ model-index:
|
|
| 28 |
type: ce-val
|
| 29 |
metrics:
|
| 30 |
- type: accuracy
|
| 31 |
-
value: 0.
|
| 32 |
name: Accuracy
|
| 33 |
- type: accuracy_threshold
|
| 34 |
-
value: -0.
|
| 35 |
name: Accuracy Threshold
|
| 36 |
- type: f1
|
| 37 |
-
value: 0.
|
| 38 |
name: F1
|
| 39 |
- type: f1_threshold
|
| 40 |
-
value: -1.
|
| 41 |
name: F1 Threshold
|
| 42 |
- type: precision
|
| 43 |
-
value: 0.
|
| 44 |
name: Precision
|
| 45 |
- type: recall
|
| 46 |
-
value: 0.
|
| 47 |
name: Recall
|
| 48 |
- type: average_precision
|
| 49 |
-
value: 0.
|
| 50 |
name: Average Precision
|
| 51 |
---
|
| 52 |
|
|
@@ -99,25 +99,25 @@ from sentence_transformers import CrossEncoder
|
|
| 99 |
model = CrossEncoder("cross_encoder_model_id")
|
| 100 |
# Get scores for pairs of inputs
|
| 101 |
pairs = [
|
| 102 |
-
['
|
| 103 |
-
['
|
| 104 |
-
['
|
| 105 |
-
['
|
| 106 |
-
['
|
| 107 |
]
|
| 108 |
scores = model.predict(pairs)
|
| 109 |
print(scores)
|
| 110 |
-
# [
|
| 111 |
|
| 112 |
# Or rank different texts based on similarity to a single text
|
| 113 |
ranks = model.rank(
|
| 114 |
-
'
|
| 115 |
[
|
| 116 |
-
'
|
| 117 |
-
'
|
| 118 |
-
'
|
| 119 |
-
'
|
| 120 |
-
'
|
| 121 |
]
|
| 122 |
)
|
| 123 |
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]
|
|
@@ -158,13 +158,13 @@ You can finetune this model on your own dataset.
|
|
| 158 |
|
| 159 |
| Metric | Value |
|
| 160 |
|:----------------------|:-----------|
|
| 161 |
-
| accuracy | 0.
|
| 162 |
-
| accuracy_threshold | -0.
|
| 163 |
-
| f1 | 0.
|
| 164 |
-
| f1_threshold | -1.
|
| 165 |
-
| precision | 0.
|
| 166 |
-
| recall | 0.
|
| 167 |
-
| **average_precision** | **0.
|
| 168 |
|
| 169 |
<!--
|
| 170 |
## Bias, Risks and Limitations
|
|
@@ -184,19 +184,19 @@ You can finetune this model on your own dataset.
|
|
| 184 |
|
| 185 |
#### Unnamed Dataset
|
| 186 |
|
| 187 |
-
* Size:
|
| 188 |
* Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
|
| 189 |
* Approximate statistics based on the first 1000 samples:
|
| 190 |
-
| | sentence_0 | sentence_1
|
| 191 |
-
|:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------|
|
| 192 |
-
| type | string | string
|
| 193 |
-
| details | <ul><li>min: 7 tokens</li><li>mean: 27.
|
| 194 |
* Samples:
|
| 195 |
-
| sentence_0
|
| 196 |
-
|:----------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------|
|
| 197 |
-
| <code>
|
| 198 |
-
| <code>
|
| 199 |
-
| <code>
|
| 200 |
* Loss: [<code>BinaryCrossEntropyLoss</code>](https://sbert.net/docs/package_reference/cross_encoder/losses.html#binarycrossentropyloss) with these parameters:
|
| 201 |
```json
|
| 202 |
{
|
|
@@ -317,11 +317,11 @@ You can finetune this model on your own dataset.
|
|
| 317 |
### Training Logs
|
| 318 |
| Epoch | Step | ce-val_average_precision |
|
| 319 |
|:-----:|:----:|:------------------------:|
|
| 320 |
-
| 1
|
| 321 |
|
| 322 |
|
| 323 |
### Training Time
|
| 324 |
-
- **Training**:
|
| 325 |
|
| 326 |
### Framework Versions
|
| 327 |
- Python: 3.12.13
|
|
|
|
| 4 |
- cross-encoder
|
| 5 |
- reranker
|
| 6 |
- generated_from_trainer
|
| 7 |
+
- dataset_size:7419
|
| 8 |
- loss:BinaryCrossEntropyLoss
|
| 9 |
base_model: cross-encoder/ms-marco-MiniLM-L6-v2
|
| 10 |
pipeline_tag: text-ranking
|
|
|
|
| 28 |
type: ce-val
|
| 29 |
metrics:
|
| 30 |
- type: accuracy
|
| 31 |
+
value: 0.816969696969697
|
| 32 |
name: Accuracy
|
| 33 |
- type: accuracy_threshold
|
| 34 |
+
value: -0.6473360061645508
|
| 35 |
name: Accuracy Threshold
|
| 36 |
- type: f1
|
| 37 |
+
value: 0.8307349665924276
|
| 38 |
name: F1
|
| 39 |
- type: f1_threshold
|
| 40 |
+
value: -1.014681339263916
|
| 41 |
name: F1 Threshold
|
| 42 |
- type: precision
|
| 43 |
+
value: 0.7690721649484537
|
| 44 |
name: Precision
|
| 45 |
- type: recall
|
| 46 |
+
value: 0.9031476997578692
|
| 47 |
name: Recall
|
| 48 |
- type: average_precision
|
| 49 |
+
value: 0.8791017476679602
|
| 50 |
name: Average Precision
|
| 51 |
---
|
| 52 |
|
|
|
|
| 99 |
model = CrossEncoder("cross_encoder_model_id")
|
| 100 |
# Get scores for pairs of inputs
|
| 101 |
pairs = [
|
| 102 |
+
['The last time the planet was even four degrees warmer, Peter Brannen points out in The Ends of the World, his new history of the planet’s major extinction events, the oceans were hundreds of feet higher.', 'It was designed by Jung Brannen Associates.'],
|
| 103 |
+
['[S]unspot activity on the surface of our star has dropped to a new low.', 'This surface activity produces starspots, which are regions of strong magnetic fields and lower than normal surface temperatures.'],
|
| 104 |
+
['More money is dedicated within the Department of Homeland Security to climate change than what\'s spent combating "Islamist terrorists radicalizing over the Internet in the United States of America."', 'According to The Washington Post, "Online recruiting has exponentially increased, with Facebook, YouTube and the increasing sophistication of people online".'],
|
| 105 |
+
['Worst-case global heating scenarios may need to be revised upwards in light of a better understanding of the role of clouds, scientists have said.', 'Climate change is more accurate scientifically to describe the various effects of greenhouse gases on the world because it includes extreme weather, storms and changes in rainfall patterns, ocean acidification and sea level.".'],
|
| 106 |
+
['Prof Adam Scaife, a climate modelling expert at the UK’s Met Office, said the evidence for a link to shrinking Arctic ice was now good: ‘The consensus points towards that being a real effect.’”', 'Category : Ceremonial officers in the United Kingdom'],
|
| 107 |
]
|
| 108 |
scores = model.predict(pairs)
|
| 109 |
print(scores)
|
| 110 |
+
# [-4.6643 0.3606 0.9265 3.4254 -5.0308]
|
| 111 |
|
| 112 |
# Or rank different texts based on similarity to a single text
|
| 113 |
ranks = model.rank(
|
| 114 |
+
'The last time the planet was even four degrees warmer, Peter Brannen points out in The Ends of the World, his new history of the planet’s major extinction events, the oceans were hundreds of feet higher.',
|
| 115 |
[
|
| 116 |
+
'It was designed by Jung Brannen Associates.',
|
| 117 |
+
'This surface activity produces starspots, which are regions of strong magnetic fields and lower than normal surface temperatures.',
|
| 118 |
+
'According to The Washington Post, "Online recruiting has exponentially increased, with Facebook, YouTube and the increasing sophistication of people online".',
|
| 119 |
+
'Climate change is more accurate scientifically to describe the various effects of greenhouse gases on the world because it includes extreme weather, storms and changes in rainfall patterns, ocean acidification and sea level.".',
|
| 120 |
+
'Category : Ceremonial officers in the United Kingdom',
|
| 121 |
]
|
| 122 |
)
|
| 123 |
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]
|
|
|
|
| 158 |
|
| 159 |
| Metric | Value |
|
| 160 |
|:----------------------|:-----------|
|
| 161 |
+
| accuracy | 0.817 |
|
| 162 |
+
| accuracy_threshold | -0.6473 |
|
| 163 |
+
| f1 | 0.8307 |
|
| 164 |
+
| f1_threshold | -1.0147 |
|
| 165 |
+
| precision | 0.7691 |
|
| 166 |
+
| recall | 0.9031 |
|
| 167 |
+
| **average_precision** | **0.8791** |
|
| 168 |
|
| 169 |
<!--
|
| 170 |
## Bias, Risks and Limitations
|
|
|
|
| 184 |
|
| 185 |
#### Unnamed Dataset
|
| 186 |
|
| 187 |
+
* Size: 7,419 training samples
|
| 188 |
* Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
|
| 189 |
* Approximate statistics based on the first 1000 samples:
|
| 190 |
+
| | sentence_0 | sentence_1 | label |
|
| 191 |
+
|:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:---------------------------------------------------------------|
|
| 192 |
+
| type | string | string | float |
|
| 193 |
+
| details | <ul><li>min: 7 tokens</li><li>mean: 27.57 tokens</li><li>max: 82 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 31.56 tokens</li><li>max: 321 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.52</li><li>max: 1.0</li></ul> |
|
| 194 |
* Samples:
|
| 195 |
+
| sentence_0 | sentence_1 | label |
|
| 196 |
+
|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------|
|
| 197 |
+
| <code>The last time the planet was even four degrees warmer, Peter Brannen points out in The Ends of the World, his new history of the planet’s major extinction events, the oceans were hundreds of feet higher.</code> | <code>It was designed by Jung Brannen Associates.</code> | <code>0.0</code> |
|
| 198 |
+
| <code>[S]unspot activity on the surface of our star has dropped to a new low.</code> | <code>This surface activity produces starspots, which are regions of strong magnetic fields and lower than normal surface temperatures.</code> | <code>1.0</code> |
|
| 199 |
+
| <code>More money is dedicated within the Department of Homeland Security to climate change than what's spent combating "Islamist terrorists radicalizing over the Internet in the United States of America."</code> | <code>According to The Washington Post, "Online recruiting has exponentially increased, with Facebook, YouTube and the increasing sophistication of people online".</code> | <code>1.0</code> |
|
| 200 |
* Loss: [<code>BinaryCrossEntropyLoss</code>](https://sbert.net/docs/package_reference/cross_encoder/losses.html#binarycrossentropyloss) with these parameters:
|
| 201 |
```json
|
| 202 |
{
|
|
|
|
| 317 |
### Training Logs
|
| 318 |
| Epoch | Step | ce-val_average_precision |
|
| 319 |
|:-----:|:----:|:------------------------:|
|
| 320 |
+
| -1 | -1 | 0.8791 |
|
| 321 |
|
| 322 |
|
| 323 |
### Training Time
|
| 324 |
+
- **Training**: 28.9 seconds
|
| 325 |
|
| 326 |
### Framework Versions
|
| 327 |
- Python: 3.12.13
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 90866404
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:63f48e7c8fd1c45a0ca9652968db194b8bfeba59c550dd8e80fd22dd4efd5b04
|
| 3 |
size 90866404
|