From 58670260a454d0ae6edc17c9cc9095c8e70d9aab Mon Sep 17 00:00:00 2001
From: jeanpoll <jean-baptiste.polle@lautorite.qc.ca>
Date: Tue, 27 Apr 2021 23:35:49 -0400
Subject: [PATCH] New improved model trained on full dataset

---
 README.md         | 37 ++++++++++++++++++++-----------------
 pytorch_model.bin |  4 ++--
 2 files changed, 22 insertions(+), 19 deletions(-)

diff --git a/README.md b/README.md
index 28e5552..87aad27 100644
--- a/README.md
+++ b/README.md
@@ -3,7 +3,7 @@ language: fr
 datasets:
 - Jean-Baptiste/wikiner_fr
 widget:
-- text: "Je m'appelle Jean-Baptiste et je vis à Paris"
+- text: "Je m'appelle jean-baptiste et je vis à montréal"
 ---
 
 # camembert-ner: model fine-tuned from camemBERT for NER task.
@@ -11,7 +11,9 @@ widget:
 ## Introduction
 
 [camembert-ner] is a NER model that was fine-tuned from camemBERT on wikiner-fr dataset.
-Model was trained on subset of wikiner-fr dataset (~36 000 sentences)
+Model was trained on wikiner-fr dataset (~170 634  sentences).
+Model was validated on emails/chat data and surperformed other models on this type of data specifically. 
+In particular the model seems to work better on entity that don't start with an upper case.
 
 
 ## How to use camembert-ner with HuggingFace
@@ -81,24 +83,25 @@ nlp("Apple est créée le 1er avril 1976 dans le garage de la maison d'enfance d
 
 Global
 ```
-'precision': 0.8830965723967158
-'recall': 0.8915789473684211
-'f1': 0.8873174883781837
+'precision': 0.8859
+'recall': 0.8971
+'f1': 0.8914
 ```
 
 By entity
 ```
-'LOC': {'precision': 0.8701754385964913,
-        'recall': 0.8878281622911695,
-        'f1': 0.8789131718842291},
-'MISC': {'precision': 0.831053901850362,
-         'recall': 0.815955766192733,
-          'f1': 0.823435631725787},
-'ORG': {'precision': 0.8620199146514936,
-        'recall': 0.8335625859697386,
-         'f1': 0.8475524475524475},
- 'PER': {'precision': 0.9367143476376246,
-         'recall': 0.9583148558758315,
-         'f1': 0.947391494958}
+'LOC': {'precision': 0.8905576596578294,
+		'recall': 0.900554675118859,
+		'f1': 0.8955282684352223},
+'MISC': {'precision': 0.8175627240143369,
+		 'recall': 0.8117437722419929,
+		 'f1': 0.8146428571428571},
+'ORG': {'precision': 0.8099480326651819,
+		'recall': 0.8265151515151515,
+		'f1': 0.8181477315335584},
+'PER': {'precision': 0.9372509960159362,
+		'recall': 0.959812321501428,
+		'f1': 0.9483975005039308}
+
  ```
 
diff --git a/pytorch_model.bin b/pytorch_model.bin
index 718630b..ab8b1ae 100644
--- a/pytorch_model.bin
+++ b/pytorch_model.bin
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55e4e2730fadcb7f4905c10d6f39abc0fe59ead7f919e5fde845f39144cc430b
-size 440227439
+oid sha256:c9f586c5bc5943992fa49fe0c0c390dace2a48288d1cec0680cd96fcd17ed037
+size 440227047