Add the pretrained weights

README.md

@@ -0,0 +1,51 @@
+---
+tags:
+- model_hub_mixin
+- pytorch_model_hub_mixin
+- address-parsing
+- deepparse
+license: mit
+datasets:
+- deepparse/worldwide-addresses
+library_name: deepparse
+---
+
+# deepparse model2vec-32m-attention
+
+A [deepparse](https://deepparse.org) address parser (Seq2Seq, with attention mechanism) using the
+[minishlab/potion-base-32M](https://huggingface.co/minishlab/potion-base-32M) Model2Vec static embeddings (512
+dimensions).
+
+```python
+from deepparse.parser import AddressParser
+
+address_parser = AddressParser(model_type="model2vec-32m", attention_mechanism=True)
+address_parser("350 rue des Lilas Ouest Québec Québec G1L 1B6")
+```
+
+Install it with `pip install deepparse[model2vec]`.
+
+## Training
+
+- Data: the [deepparse/worldwide-addresses](https://huggingface.co/datasets/deepparse/worldwide-addresses) dataset
+  (20 countries, 100,000 clean and 50,000 incomplete addresses per country), 90% for training and 10% for validation.
+  `Suburb` and `District` are mapped to `Municipality`, and the `County` and `Country` words are removed.
+- Architecture: encoder and decoder hidden size 1024, one layer each, with an attention mechanism.
+- Selection: the hidden size (128, 256, 512 or 1024) with the lowest mean validation loss over 5 seeds, then the
+  seed with the lowest validation loss (seed 43, best epoch 39).
+
+## Results
+
+| | Accuracy (%) |
+|---|---|
+| Test, this model | 98.42 |
+| Test, mean ± std over the 5 seeds of this configuration | 97.89 ± 0.58 |
+| Validation, this model | 98.65 |
+
+The accuracy is the per-word tag accuracy on the test split of the rebuilt worldwide-addresses data.
+
+## Limitations
+
+The source data has no `Orientation` nor `GeneralDelivery` tags: the model has these outputs but never predicts them
+(the street orientation, e.g. `Ouest` or `NW`, is part of the `StreetName`). South Korea has almost only incomplete
+addresses.

config.json

@@ -0,0 +1,9 @@
+{
+  "attention_mechanism": true,
+  "decoder_hidden_size": 1024,
+  "decoder_num_layers": 1,
+  "encoder_hidden_size": 1024,
+  "encoder_num_layers": 1,
+  "input_size": 512,
+  "output_size": 9
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fd0d5e0e6867a86be45662f9e38409019546699f9895e9f9c18b6ca29834641
+size 67241476