Text Classification
Transformers
Safetensors
prompt-complexity
feature-extraction
regression
prompt
complexity-estimation
semantic-routing
llm-routing
custom_code
Instructions to use ilya-kolchinsky/PromptComplexityEstimator with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ilya-kolchinsky/PromptComplexityEstimator with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="ilya-kolchinsky/PromptComplexityEstimator", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("ilya-kolchinsky/PromptComplexityEstimator", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -107,7 +107,7 @@ def route_prompt(prompt: str) -> str:
|
|
| 107 |
### Model
|
| 108 |
- **Backbone encoder:** microsoft/deberta-v3-base
|
| 109 |
- Mask-aware **mean pooling** over token embeddings + **LayerNorm**
|
| 110 |
-
- **Regression head:**
|
| 111 |
- **Max input length:** 512 tokens
|
| 112 |
- The model outputs a bounded score in [0, 1]. In the examples below, the score is read from `outputs.logits` (shape `[batch, 1]`).
|
| 113 |
|
|
|
|
| 107 |
### Model
|
| 108 |
- **Backbone encoder:** microsoft/deberta-v3-base
|
| 109 |
- Mask-aware **mean pooling** over token embeddings + **LayerNorm**
|
| 110 |
+
- **Regression head:** Linear → ReLU → Linear → Sigmoid
|
| 111 |
- **Max input length:** 512 tokens
|
| 112 |
- The model outputs a bounded score in [0, 1]. In the examples below, the score is read from `outputs.logits` (shape `[batch, 1]`).
|
| 113 |
|