Update Readme

README.md

@@ -1,21 +1,36 @@
-# PyLaia Rimes
+---
+library_name: PyLaia
+license: mit
+tags:
+- PyLaia
+- PyTorch
+- Handwritten text recognition
+metrics:
+- CER
+- WER
+language:
+- fr
+---
 
-## Datasets
+# French handwritten text recognition
 
-Trained on text-lines from the [Rimes 2011 dataset](https://teklia.com/research/rimes-database/).
+This model performs Handwritten Text Recognition in French.
 
-| split  | N lines |
-|--------|--------:|
-| train  | 10,188  |
-| val    |  1,138  |
-| test   |    778  |
+## Model description
 
-## Results
+The model has been trained using the PyLaia library on the [RIMES](https://teklia.com/research/rimes-database/) dataset.
+Training images were resized with a fixed height of 128 pixels, keeping the original aspect ratio.
 
-* Fixed line height: 128 pixels
-* Language model: 6-gram character model trained on the training set with KenLM
 
-| Model             			  | val CER | test CER | val WER | test WER |
-|:--------------------------------|--------:|---------:|--------:|---------:|
-| Model without LM 	              | 4.55  	| 4.53 	   | 14.39   | 15.06    |
-| Model with LM (`weight = 1.5`)  | 3.68    | 3.47 	   | 10.01 	 | 10.20    |
+## Evaluation results
+
+The model achieves the following results:
+
+| set   | Language model | CER (%)    | WER (%) | N lines   |
+|:------|:---------------|:----------:|:-------:|----------:|
+| test  | no             | 4.53       | 15.06   |      778  |
+| test  | yes            | 3.47       | 10.20   |      778  |
+
+## How to use
+
+Please refer to the [documentation](https://atr.pages.teklia.com/pylaia/).