Text Classification
Transformers
ONNX
Safetensors
English
distilbert
intent-classification
multitask
iab
conversational-ai
adtech
calibrated-confidence
text-embeddings-inference
Instructions to use admesh/agentic-intent-classifier with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use admesh/agentic-intent-classifier with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="admesh/agentic-intent-classifier")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("admesh/agentic-intent-classifier", dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "accepted_accuracy": 0.943, | |
| "accepted_coverage": 1.0, | |
| "accuracy": 0.943, | |
| "count": 3282, | |
| "dataset_path": "/content/agentic-intent-classifier/data/iab/test.jsonl", | |
| "fallback_rate": 0.0, | |
| "head": "iab_content", | |
| "macro_f1": 0.911, | |
| "primary_source": "supervised_classifier", | |
| "suite": "test", | |
| "tier_metrics": { | |
| "average_prediction_depth": 2.213, | |
| "error_buckets": { | |
| "exact_match": 3095, | |
| "parent_safe_stop": 45, | |
| "right_tier1_wrong_tier2": 41, | |
| "wrong_deep_leaf": 72, | |
| "wrong_tier1": 29 | |
| }, | |
| "exact_path_accuracy": 0.943, | |
| "parent_safe_accuracy": 0.958, | |
| "tier1_accuracy": 0.9912, | |
| "tier2_accuracy": 0.9776, | |
| "tier3_accuracy": 0.9078, | |
| "tier4_accuracy": 0.7 | |
| }, | |
| "view_metrics": { | |
| "classifier": { | |
| "average_prediction_depth": 2.213, | |
| "error_buckets": { | |
| "exact_match": 3052, | |
| "parent_safe_stop": 44, | |
| "right_tier1_wrong_tier2": 53, | |
| "wrong_deep_leaf": 104, | |
| "wrong_tier1": 29 | |
| }, | |
| "exact_path_accuracy": 0.9299, | |
| "parent_safe_accuracy": 0.9445, | |
| "tier1_accuracy": 0.9912, | |
| "tier2_accuracy": 0.9734, | |
| "tier3_accuracy": 0.8725, | |
| "tier4_accuracy": 0.5 | |
| }, | |
| "combined_path": { | |
| "count": 3282, | |
| "max_combined_rows": 500, | |
| "reason": "dataset_too_large_for_combined_view", | |
| "skipped": true | |
| }, | |
| "disagreements": { | |
| "count": 3282, | |
| "max_combined_rows": 500, | |
| "reason": "dataset_too_large_for_combined_view", | |
| "skipped": true | |
| }, | |
| "shadow_embedding_retrieval": { | |
| "hint": "Set IAB_EVAL_INCLUDE_SHADOW_RETRIEVAL=1 to run shadow embedding retrieval (downloads/loads gte-Qwen2 when index is present).", | |
| "reason": "disabled_by_default", | |
| "skipped": true | |
| } | |
| } | |
| } | |