Text Classification
Transformers
Laya
English
multilingual
typed-decisions
non-autoregressive
axera
ax650
Instructions to use AXERA-TECH/Laya with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AXERA-TECH/Laya with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="AXERA-TECH/Laya")# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("AXERA-TECH/Laya", device_map="auto") - Laya
How to use AXERA-TECH/Laya with Laya:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
yongqiang commited on
Commit ·
ac8a43d
1
Parent(s): 035fa9a
Correct AX637 decision agreement
Browse files- AX637_BACKEND/README.md +2 -2
AX637_BACKEND/README.md
CHANGED
|
@@ -456,9 +456,9 @@ The package includes the exact requests and complete JSON outputs used for board
|
|
| 456 |
| --- | --- | --- | ---: |
|
| 457 |
| `english` | Duplicate charge, refund request, cancellation threat | `billing` 0.9690; refund 0.8548; churn 0.8346 | 63/64 |
|
| 458 |
| `multilingual` | Chinese duplicate charge and cancellation intent | `billing` 1.0000; refund 0.9936; churn 0.9307 | 60/64 |
|
| 459 |
-
| `typed-decisions` | Active production API key exposed publicly | `contain` 0.4082; severity 1.7202; containment 0.6667 |
|
| 460 |
|
| 461 |
-
Differing cases occur at low-confidence boundaries, such as close choice labels, adjacent score levels, or noul probabilities near 0.5. Treat close probabilities as uncertain instead of relying only on the winning label.
|
| 462 |
|
| 463 |
## Packaged Graph Constraints
|
| 464 |
|
|
|
|
| 456 |
| --- | --- | --- | ---: |
|
| 457 |
| `english` | Duplicate charge, refund request, cancellation threat | `billing` 0.9690; refund 0.8548; churn 0.8346 | 63/64 |
|
| 458 |
| `multilingual` | Chinese duplicate charge and cancellation intent | `billing` 1.0000; refund 0.9936; churn 0.9307 | 60/64 |
|
| 459 |
+
| `typed-decisions` | Active production API key exposed publicly | `contain` 0.4082; severity 1.7202; containment 0.6667 | 62/64 |
|
| 460 |
|
| 461 |
+
Differing cases occur at low-confidence boundaries, such as close choice labels, adjacent score levels, or noul probabilities near 0.5. For `noul`, agreement uses the inference rule `probability >= 0.5`; an exact 0.5 is counted as true. Treat close probabilities as uncertain instead of relying only on the winning label.
|
| 462 |
|
| 463 |
## Packaged Graph Constraints
|
| 464 |
|