Token Classification
Transformers
PyTorch
English
Hindi
ner
address-parsing
indian-addresses
bert
crf
Eval Results (legacy)
Instructions to use howdoiuse-keyboard/indian-address-parser-model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use howdoiuse-keyboard/indian-address-parser-model with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="howdoiuse-keyboard/indian-address-parser-model")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("howdoiuse-keyboard/indian-address-parser-model", dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "test_f1": 0.9430501930501931, | |
| "test_precision": 0.9439613526570049, | |
| "test_recall": 0.9421407907425265, | |
| "test_report": " precision recall f1-score support\n\n AREA 1.00 1.00 1.00 67\n BLOCK 0.97 0.97 0.97 63\n CITY 0.99 1.00 1.00 142\n COLONY 0.97 0.97 0.97 77\n FLOOR 0.94 0.93 0.93 67\n GALI 1.00 1.00 1.00 68\nHOUSE_NUMBER 0.99 0.97 0.98 73\n KHASRA 0.87 0.87 0.87 63\n PINCODE 1.00 1.00 1.00 138\n PLOT 0.38 0.37 0.37 62\n SECTOR 1.00 1.00 1.00 63\n STATE 1.00 0.98 0.99 63\n SUBAREA 0.98 0.99 0.98 91\n\n micro avg 0.94 0.94 0.94 1037\n macro avg 0.93 0.93 0.93 1037\nweighted avg 0.94 0.94 0.94 1037\n" | |
| } |