fgiauna's picture
update model card README.md
142422f
|
raw
history blame
8.36 kB
metadata
license: mit
tags:
  - generated_from_trainer
model-index:
  - name: camembert-ner-finetuned-jul
    results: []

camembert-ner-finetuned-jul

This model is a fine-tuned version of Jean-Baptiste/camembert-ner on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6420
  • Loc: {'precision': 0.706140350877193, 'recall': 0.4586894586894587, 'f1': 0.5561312607944733, 'number': 1053}
  • Misc: {'precision': 0.037747920665387076, 'recall': 0.6820809248554913, 'f1': 0.07153682934222491, 'number': 173}
  • Org: {'precision': 0.6212424849699398, 'recall': 0.3125, 'f1': 0.4158283031522468, 'number': 992}
  • Per: {'precision': 0.5895458440445587, 'recall': 0.67384916748286, 'f1': 0.6288848263254113, 'number': 1021}
  • Overall Precision: 0.2920
  • Overall Recall: 0.4937
  • Overall F1: 0.3670
  • Overall Accuracy: 0.9079

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Loc Misc Org Per Overall Precision Overall Recall Overall F1 Overall Accuracy
0.125 1.0 2033 0.4529 {'precision': 0.7350565428109854, 'recall': 0.43209876543209874, 'f1': 0.5442583732057417, 'number': 1053} {'precision': 0.03518518518518519, 'recall': 0.6589595375722543, 'f1': 0.06680339876941108, 'number': 173} {'precision': 0.6746666666666666, 'recall': 0.2550403225806452, 'f1': 0.37015362106803223, 'number': 992} {'precision': 0.6089686098654709, 'recall': 0.6650342801175319, 'f1': 0.6357677902621722, 'number': 1021} 0.2806 0.4634 0.3496 0.8956
0.0809 2.0 4066 0.3927 {'precision': 0.733652312599681, 'recall': 0.4368471035137702, 'f1': 0.5476190476190477, 'number': 1053} {'precision': 0.03864382063434196, 'recall': 0.6127167630057804, 'f1': 0.07270233196159122, 'number': 173} {'precision': 0.6554307116104869, 'recall': 0.3528225806451613, 'f1': 0.45871559633027525, 'number': 992} {'precision': 0.561034761519806, 'recall': 0.6797257590597453, 'f1': 0.6147032772364924, 'number': 1021} 0.3132 0.4971 0.3842 0.9113
0.0609 3.0 6099 0.4609 {'precision': 0.6685714285714286, 'recall': 0.4444444444444444, 'f1': 0.5339418140330862, 'number': 1053} {'precision': 0.03744567034436643, 'recall': 0.6473988439306358, 'f1': 0.07079646017699115, 'number': 173} {'precision': 0.656964656964657, 'recall': 0.3185483870967742, 'f1': 0.4290563475899525, 'number': 992} {'precision': 0.5294953802416489, 'recall': 0.7296767874632712, 'f1': 0.6136738056013179, 'number': 1021} 0.2941 0.5066 0.3722 0.9045
0.0423 4.0 8132 0.4695 {'precision': 0.7260940032414911, 'recall': 0.4254510921177588, 'f1': 0.5365269461077844, 'number': 1053} {'precision': 0.04154204054503157, 'recall': 0.7225433526011561, 'f1': 0.07856693903205532, 'number': 173} {'precision': 0.6585903083700441, 'recall': 0.3014112903225806, 'f1': 0.4135546334716459, 'number': 992} {'precision': 0.552366175329713, 'recall': 0.6973555337904016, 'f1': 0.6164502164502165, 'number': 1021} 0.2950 0.4890 0.3680 0.9094
0.0354 5.0 10165 0.5324 {'precision': 0.6918518518518518, 'recall': 0.44349477682811017, 'f1': 0.5405092592592593, 'number': 1053} {'precision': 0.03929273084479371, 'recall': 0.6936416184971098, 'f1': 0.07437248218159281, 'number': 173} {'precision': 0.6088794926004228, 'recall': 0.2903225806451613, 'f1': 0.39317406143344713, 'number': 992} {'precision': 0.5702054794520548, 'recall': 0.6523016650342801, 'f1': 0.6084970306075833, 'number': 1021} 0.2870 0.4758 0.3580 0.9067
0.0254 6.0 12198 0.5827 {'precision': 0.684593023255814, 'recall': 0.4472934472934473, 'f1': 0.5410683515221136, 'number': 1053} {'precision': 0.031561996779388084, 'recall': 0.5664739884393064, 'f1': 0.05979255643685173, 'number': 173} {'precision': 0.6134301270417423, 'recall': 0.3407258064516129, 'f1': 0.43810758263123784, 'number': 992} {'precision': 0.5973684210526315, 'recall': 0.6669931439764937, 'f1': 0.6302637667746414, 'number': 1021} 0.2896 0.4903 0.3641 0.9046
0.0206 7.0 14231 0.5821 {'precision': 0.6886657101865137, 'recall': 0.45584045584045585, 'f1': 0.5485714285714286, 'number': 1053} {'precision': 0.03427093436792758, 'recall': 0.6127167630057804, 'f1': 0.06491120636864667, 'number': 173} {'precision': 0.5944055944055944, 'recall': 0.34274193548387094, 'f1': 0.43478260869565216, 'number': 992} {'precision': 0.5777964676198486, 'recall': 0.672869735553379, 'f1': 0.6217194570135747, 'number': 1021} 0.2906 0.4980 0.3670 0.9063
0.0147 8.0 16264 0.6210 {'precision': 0.7218045112781954, 'recall': 0.45584045584045585, 'f1': 0.5587892898719441, 'number': 1053} {'precision': 0.03523542251325226, 'recall': 0.653179190751445, 'f1': 0.06686390532544378, 'number': 173} {'precision': 0.6188524590163934, 'recall': 0.30443548387096775, 'f1': 0.40810810810810816, 'number': 992} {'precision': 0.6246362754607178, 'recall': 0.6307541625857003, 'f1': 0.6276803118908383, 'number': 1021} 0.2855 0.4751 0.3567 0.9047
0.0123 9.0 18297 0.6424 {'precision': 0.7036496350364964, 'recall': 0.4577397910731244, 'f1': 0.5546605293440737, 'number': 1053} {'precision': 0.04029899252518687, 'recall': 0.7167630057803468, 'f1': 0.07630769230769231, 'number': 173} {'precision': 0.6023391812865497, 'recall': 0.31149193548387094, 'f1': 0.41063122923588036, 'number': 992} {'precision': 0.5862068965517241, 'recall': 0.682664054848188, 'f1': 0.6307692307692307, 'number': 1021} 0.2950 0.4977 0.3704 0.9074
0.0093 10.0 20330 0.6420 {'precision': 0.706140350877193, 'recall': 0.4586894586894587, 'f1': 0.5561312607944733, 'number': 1053} {'precision': 0.037747920665387076, 'recall': 0.6820809248554913, 'f1': 0.07153682934222491, 'number': 173} {'precision': 0.6212424849699398, 'recall': 0.3125, 'f1': 0.4158283031522468, 'number': 992} {'precision': 0.5895458440445587, 'recall': 0.67384916748286, 'f1': 0.6288848263254113, 'number': 1021} 0.2920 0.4937 0.3670 0.9079

Framework versions

  • Transformers 4.29.0
  • Pytorch 2.0.0+cu118
  • Datasets 2.12.0
  • Tokenizers 0.13.3