Adding ONNX file of this model (#4)

- Adding ONNX file of this model (d2d1ee3b883b461de7eeb2500ab6051558aa40f2)

.gitattributes

@@ -15,3 +15,4 @@
 *.pt filter=lfs diff=lfs merge=lfs -text
 *.pth filter=lfs diff=lfs merge=lfs -text
 pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
+model.safetensors filter=lfs diff=lfs merge=lfs -text

README.md

@@ -4,6 +4,8 @@ datasets:
 - Jean-Baptiste/wikiner_fr
 widget:
 - text: "Je m'appelle jean-baptiste et je vis à montréal"
+- text: "george washington est allé à washington"
+license: mit
 ---
 
 # camembert-ner: model fine-tuned from camemBERT for NER task.
@@ -12,9 +14,20 @@ widget:
 
 [camembert-ner] is a NER model that was fine-tuned from camemBERT on wikiner-fr dataset.
 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. 
+Model was validated on emails/chat data and overperformed 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.
 
+## Training data
+Training data was classified as follow:
+
+Abbreviation|Description
+-|-
+O |Outside of a named entity
+MISC |Miscellaneous entity
+PER |Person’s name
+ORG |Organization
+LOC |Location
+
 
 ## How to use camembert-ner with HuggingFace
 
@@ -31,7 +44,7 @@ model = AutoModelForTokenClassification.from_pretrained("Jean-Baptiste/camembert
 
 from transformers import pipeline
 
-nlp = pipeline('ner', model=model, tokenizer=tokenizer, grouped_entities=True)
+nlp = pipeline('ner', model=model, tokenizer=tokenizer, aggregation_strategy="simple")
 nlp("Apple est créée le 1er avril 1976 dans le garage de la maison d'enfance de Steve Jobs à Los Altos en Californie par Steve Jobs, Steve Wozniak et Ronald Wayne14, puis constituée sous forme de société le 3 janvier 1977 à l'origine sous le nom d'Apple Computer, mais pour ses 30 ans et pour refléter la diversification de ses produits, le mot « computer » est retiré le 9 janvier 2015.")
 
 
@@ -81,27 +94,23 @@ nlp("Apple est créée le 1er avril 1976 dans le garage de la maison d'enfance d
 
 ## Model performances (metric: seqeval)
 
-Global
+Overall
-```
+
-'precision': 0.8859
+precision|recall|f1
-'recall': 0.8971
+-|-|-
-'f1': 0.8914
+0.8859|0.8971|0.8914
-```
 
 By entity
-```
-'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}
 
- ```
+entity|precision|recall|f1
+-|-|-|-
+PER|0.9372|0.9598|0.9483 
+ORG|0.8099|0.8265|0.8181
+LOC|0.8905|0.9005|0.8955
+MISC|0.8175|0.8117|0.8146
 
+
+
+
+For those who could be interested, here is a short article on how I used the results of this model to train a LSTM model for signature detection in emails:
+https://medium.com/@jean-baptiste.polle/lstm-model-for-email-signature-detection-8e990384fefa

config.json

@@ -12,19 +12,19 @@
   "hidden_size": 768,
   "id2label": {
     "0": "O",
-    "1": "LOC",
+    "1": "I-LOC",
-    "2": "PER",
+    "2": "I-PER",
-    "3": "MISC",
+    "3": "I-MISC",
-    "4": "ORG"
+    "4": "I-ORG"
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "LOC": 1,
+    "I-LOC": 1,
-    "MISC": 3,
+    "I-MISC": 3,
     "O": 0,
-    "ORG": 4,
+    "I-ORG": 4,
-    "PER": 2
+    "I-PER": 2
   },
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 514,