Hugging Face's logo

AntoineBourgois
/
BookNLP-fr_NER_camembert-large_FAC_GPE_LOC_PER_TIME_VEH

Token Classification
French
NER
camembert
literary-texts
nested-entities
BookNLP-fr
Model card Files Community
AntoineBourgois commited on
Commit
931af1d
·
verified ·
1 Parent(s): 1316269

Upload 3 files

Browse files
Files changed (2) hide show
  1. JCLS_model_card.md +14 -14
  2. README.md +48 -48
JCLS_model_card.md CHANGED
@@ -31,14 +31,14 @@ The predicted entities are:
31
  ## MODEL PERFORMANCES (LOOCV):
32
  | NER_tag | precision | recall | f1_score | support | support % |
33
  |-----------|-------------|----------|------------|-----------|-------------|
34
- | PER | 92.97% | 96.25% | 94.58% | 4,162 | 86.12% |
35
- | FAC | 76.58% | 75.89% | 76.23% | 224 | 4.63% |
36
- | TIME | 66.97% | 69.48% | 68.20% | 213 | 4.41% |
37
- | LOC | 70.00% | 57.27% | 63.00% | 110 | 2.28% |
38
- | GPE | 80.65% | 78.12% | 79.37% | 64 | 1.32% |
39
- | VEH | 57.75% | 68.33% | 62.60% | 60 | 1.24% |
40
- | micro_avg | 89.94% | 92.65% | 91.25% | 4,833 | 100.00% |
41
- | macro_avg | 74.15% | 74.23% | 74.00% | 4,833 | 100.00% |
42
 
43
  ## TRAINING PARAMETERS:
44
  - Entities types: ['PER', 'LOC', 'FAC', 'TIME', 'VEH', 'GPE']
@@ -111,12 +111,12 @@ Model Output: BIOES labels sequence
111
  | Gold Labels | PER | FAC | TIME | LOC | GPE | VEH | O | support |
112
  |---------------|-------|-------|--------|-------|-------|-------|-----|-----------|
113
  | PER | 4,006 | 0 | 2 | 1 | 1 | 3 | 149 | 4,162 |
114
- | FAC | 8 | 170 | 0 | 2 | 0 | 1 | 43 | 224 |
115
- | TIME | 1 | 0 | 148 | 0 | 0 | 0 | 64 | 213 |
116
- | LOC | 0 | 2 | 0 | 63 | 6 | 0 | 39 | 110 |
117
- | GPE | 2 | 1 | 0 | 3 | 50 | 0 | 8 | 64 |
118
- | VEH | 3 | 0 | 0 | 0 | 0 | 41 | 16 | 60 |
119
- | O | 287 | 49 | 70 | 21 | 5 | 26 | 0 | 458 |
120
 
121
  ## CONTACT:
122
  mail: antoine [dot] bourgois [at] protonmail [dot] com
 
31
  ## MODEL PERFORMANCES (LOOCV):
32
  | NER_tag | precision | recall | f1_score | support | support % |
33
  |-----------|-------------|----------|------------|-----------|-------------|
34
+ | PER | 92.82% | 96.25% | 94.50% | 4,162 | 86.12% |
35
+ | FAC | 74.78% | 75.45% | 75.11% | 224 | 4.63% |
36
+ | TIME | 63.48% | 68.54% | 65.91% | 213 | 4.41% |
37
+ | LOC | 72.22% | 59.09% | 65.00% | 110 | 2.28% |
38
+ | GPE | 81.36% | 75.00% | 78.05% | 64 | 1.32% |
39
+ | VEH | 56.06% | 61.67% | 58.73% | 60 | 1.24% |
40
+ | micro_avg | 89.61% | 92.51% | 91.01% | 4,833 | 100.00% |
41
+ | macro_avg | 73.45% | 72.67% | 72.88% | 4,833 | 100.00% |
42
 
43
  ## TRAINING PARAMETERS:
44
  - Entities types: ['PER', 'LOC', 'FAC', 'TIME', 'VEH', 'GPE']
 
111
  | Gold Labels | PER | FAC | TIME | LOC | GPE | VEH | O | support |
112
  |---------------|-------|-------|--------|-------|-------|-------|-----|-----------|
113
  | PER | 4,006 | 0 | 2 | 1 | 1 | 3 | 149 | 4,162 |
114
+ | FAC | 7 | 169 | 0 | 2 | 0 | 0 | 46 | 224 |
115
+ | TIME | 1 | 0 | 146 | 0 | 0 | 0 | 66 | 213 |
116
+ | LOC | 1 | 3 | 0 | 65 | 6 | 0 | 35 | 110 |
117
+ | GPE | 2 | 3 | 0 | 3 | 48 | 0 | 8 | 64 |
118
+ | VEH | 5 | 1 | 0 | 0 | 0 | 37 | 17 | 60 |
119
+ | O | 292 | 50 | 81 | 19 | 4 | 26 | 0 | 472 |
120
 
121
  ## CONTACT:
122
  mail: antoine [dot] bourgois [at] protonmail [dot] com
README.md CHANGED
@@ -31,14 +31,14 @@ The predicted entities are:
31
  ## MODEL PERFORMANCES (LOOCV):
32
  | NER_tag | precision | recall | f1_score | support | support % |
33
  |-----------|-------------|----------|------------|-----------|-------------|
34
- | PER | 92.78% | 94.29% | 93.53% | 10,354 | 86.92% |
35
- | FAC | 69.81% | 69.92% | 69.87% | 635 | 5.33% |
36
- | TIME | 64.21% | 62.12% | 63.15% | 462 | 3.88% |
37
- | LOC | 63.50% | 46.28% | 53.54% | 188 | 1.58% |
38
- | GPE | 79.86% | 74.68% | 77.18% | 154 | 1.29% |
39
- | VEH | 61.82% | 57.14% | 59.39% | 119 | 1.00% |
40
- | micro_avg | 89.51% | 90.36% | 89.91% | 11,912 | 100.00% |
41
- | macro_avg | 72.00% | 67.40% | 69.44% | 11,912 | 100.00% |
42
 
43
  ## TRAINING PARAMETERS:
44
  - Entities types: ['PER', 'LOC', 'FAC', 'TIME', 'VEH', 'GPE']
@@ -75,48 +75,48 @@ Model Output: BIOES labels sequence
75
  *** IN CONSTRUCTION ***
76
 
77
  ## TRAINING CORPUS:
78
- | | Document | Tokens Count | Is included in model eval |
79
- |----|---------------------------------------------------------------------------------|----------------|-----------------------------------|
80
- | 0 | 1830_Balzac-Honoré-de_La-maison-du-chat-qui-pelote | 24,776 tokens | True |
81
- | 1 | 1830_Balzac-Honoré-de_Sarrasine | 15,408 tokens | True |
82
- | 2 | 1836_Gautier-Théophile_La-morte-amoureuse | 14,293 tokens | True |
83
- | 3 | 1837_Balzac-Honoré-de_La-maison-Nucingen | 30,034 tokens | False |
84
- | 4 | 1841_Sand-George_Pauline | 12,398 tokens | False |
85
- | 5 | 1856_Cousin-Victor_Madame-de-Hautefort | 11,768 tokens | False |
86
- | 6 | 1863_Gautier-Théophile_Le-capitaine-Fracasse | 11,848 tokens | False |
87
- | 7 | 1873_Zola-Émile_Le-ventre-de-Paris | 12,613 tokens | False |
88
- | 8 | 1881_Flaubert-Gustave_Bouvard-et-Pécuchet | 12,308 tokens | False |
89
- | 9 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-La-buche | 2,267 tokens | False |
90
- | 10 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-La-relique | 2,041 tokens | False |
91
- | 11 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-La-rouille | 2,949 tokens | True |
92
- | 12 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Madame-Baptiste | 2,578 tokens | True |
93
- | 13 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Marocca | 4,078 tokens | False |
94
- | 14 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-A-cheval | 2,878 tokens | False |
95
- | 15 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Fou | 1,905 tokens | False |
96
- | 16 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Mademoiselle-Fifi | 5,439 tokens | True |
97
- | 17 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Reveil | 2,159 tokens | False |
98
- | 18 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Un-reveillon | 2,364 tokens | False |
99
- | 19 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Une-ruse | 2,469 tokens | False |
100
- | 20 | 1901_Achard-Lucie_Rosalie-de-Constant-sa-famille-et-ses-amis | 12,775 tokens | False |
101
- | 21 | 1903_Conan-Laure_Élisabeth-Seton | 13,046 tokens | False |
102
- | 22 | 1904-1912_Rolland-Romain_Jean-Christophe(1) | 10,982 tokens | True |
103
- | 23 | 1904-1912_Rolland-Romain_Jean-Christophe(2) | 10,305 tokens | False |
104
- | 24 | 1917_Bourgeois-Adèle_Némoville | 12,468 tokens | False |
105
- | 25 | 1923_Radiguet-Raymond_Le-diable-au-corps | 14,850 tokens | False |
106
- | 26 | 1926_Audoux-Marguerite_De-la-ville-au-moulin | 12,144 tokens | True |
107
- | 27 | 1937_Audoux-Marguerite_Douce-Lumière | 12,346 tokens | False |
108
- | 28 | TOTAL | 275,489 tokens | 8 files used for cross-validation |
109
 
110
  ## PREDICTIONS CONFUSION MATRIX:
111
- | Gold Labels | PER | FAC | TIME | LOC | GPE | VEH | O | support |
112
- |---------------|-------|-------|--------|-------|-------|-------|-----|-----------|
113
- | PER | 9,763 | 3 | 6 | 1 | 1 | 6 | 574 | 10,354 |
114
- | FAC | 27 | 444 | 1 | 4 | 4 | 1 | 154 | 635 |
115
- | TIME | 1 | 0 | 287 | 0 | 0 | 0 | 174 | 462 |
116
- | LOC | 1 | 13 | 0 | 87 | 11 | 0 | 76 | 188 |
117
- | GPE | 3 | 2 | 1 | 8 | 115 | 0 | 25 | 154 |
118
- | VEH | 12 | 1 | 0 | 0 | 0 | 68 | 38 | 119 |
119
- | O | 709 | 168 | 151 | 37 | 13 | 35 | 0 | 1,113 |
120
 
121
  ## CONTACT:
122
  mail: antoine [dot] bourgois [at] protonmail [dot] com
 
31
  ## MODEL PERFORMANCES (LOOCV):
32
  | NER_tag | precision | recall | f1_score | support | support % |
33
  |-----------|-------------|----------|------------|-----------|-------------|
34
+ | PER | 92.46% | 93.71% | 93.08% | 32,204 | 84.13% |
35
+ | FAC | 70.63% | 70.94% | 70.78% | 2,295 | 6.00% |
36
+ | TIME | 58.66% | 57.75% | 58.20% | 1,671 | 4.37% |
37
+ | GPE | 77.64% | 77.37% | 77.50% | 866 | 2.26% |
38
+ | LOC | 62.96% | 45.71% | 52.97% | 781 | 2.04% |
39
+ | VEH | 63.43% | 47.95% | 54.61% | 463 | 1.21% |
40
+ | micro_avg | 88.39% | 88.87% | 88.58% | 38,280 | 100.00% |
41
+ | macro_avg | 70.96% | 65.57% | 67.86% | 38,280 | 100.00% |
42
 
43
  ## TRAINING PARAMETERS:
44
  - Entities types: ['PER', 'LOC', 'FAC', 'TIME', 'VEH', 'GPE']
 
75
  *** IN CONSTRUCTION ***
76
 
77
  ## TRAINING CORPUS:
78
+ | | Document | Tokens Count | Is included in model eval |
79
+ |----|---------------------------------------------------------------------------------|----------------|------------------------------------|
80
+ | 0 | 1830_Balzac-Honoré-de_La-maison-du-chat-qui-pelote | 24,776 tokens | True |
81
+ | 1 | 1830_Balzac-Honoré-de_Sarrasine | 15,408 tokens | True |
82
+ | 2 | 1836_Gautier-Théophile_La-morte-amoureuse | 14,293 tokens | True |
83
+ | 3 | 1837_Balzac-Honoré-de_La-maison-Nucingen | 30,034 tokens | True |
84
+ | 4 | 1841_Sand-George_Pauline | 12,398 tokens | True |
85
+ | 5 | 1856_Cousin-Victor_Madame-de-Hautefort | 11,768 tokens | True |
86
+ | 6 | 1863_Gautier-Théophile_Le-capitaine-Fracasse | 11,848 tokens | True |
87
+ | 7 | 1873_Zola-Émile_Le-ventre-de-Paris | 12,613 tokens | True |
88
+ | 8 | 1881_Flaubert-Gustave_Bouvard-et-Pécuchet | 12,308 tokens | True |
89
+ | 9 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-La-buche | 2,267 tokens | True |
90
+ | 10 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-La-relique | 2,041 tokens | True |
91
+ | 11 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-La-rouille | 2,949 tokens | True |
92
+ | 12 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Madame-Baptiste | 2,578 tokens | True |
93
+ | 13 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Marocca | 4,078 tokens | True |
94
+ | 14 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-A-cheval | 2,878 tokens | True |
95
+ | 15 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Fou | 1,905 tokens | True |
96
+ | 16 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Mademoiselle-Fifi | 5,439 tokens | True |
97
+ | 17 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Reveil | 2,159 tokens | True |
98
+ | 18 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Un-reveillon | 2,364 tokens | True |
99
+ | 19 | 1882-1883_Maupassant-Guy-de_Mademoiselle-Fifi-Nouveaux-contes-Une-ruse | 2,469 tokens | True |
100
+ | 20 | 1901_Achard-Lucie_Rosalie-de-Constant-sa-famille-et-ses-amis | 12,775 tokens | True |
101
+ | 21 | 1903_Conan-Laure_Élisabeth-Seton | 13,046 tokens | True |
102
+ | 22 | 1904-1912_Rolland-Romain_Jean-Christophe(1) | 10,982 tokens | True |
103
+ | 23 | 1904-1912_Rolland-Romain_Jean-Christophe(2) | 10,305 tokens | True |
104
+ | 24 | 1917_Bourgeois-Adèle_Némoville | 12,468 tokens | True |
105
+ | 25 | 1923_Radiguet-Raymond_Le-diable-au-corps | 14,850 tokens | True |
106
+ | 26 | 1926_Audoux-Marguerite_De-la-ville-au-moulin | 12,144 tokens | True |
107
+ | 27 | 1937_Audoux-Marguerite_Douce-Lumière | 12,346 tokens | True |
108
+ | 28 | TOTAL | 275,489 tokens | 28 files used for cross-validation |
109
 
110
  ## PREDICTIONS CONFUSION MATRIX:
111
+ | Gold Labels | PER | FAC | TIME | GPE | LOC | VEH | O | support |
112
+ |---------------|--------|-------|--------|-------|-------|-------|-------|-----------|
113
+ | PER | 30,177 | 28 | 14 | 7 | 7 | 31 | 1,940 | 32,204 |
114
+ | FAC | 42 | 1,628 | 1 | 22 | 17 | 1 | 584 | 2,295 |
115
+ | TIME | 8 | 1 | 965 | 1 | 1 | 0 | 695 | 1,671 |
116
+ | GPE | 13 | 31 | 2 | 670 | 31 | 0 | 119 | 866 |
117
+ | LOC | 8 | 64 | 1 | 56 | 357 | 0 | 295 | 781 |
118
+ | VEH | 54 | 8 | 0 | 0 | 0 | 222 | 179 | 463 |
119
+ | O | 2,285 | 524 | 661 | 100 | 150 | 96 | 0 | 3,816 |
120
 
121
  ## CONTACT:
122
  mail: antoine [dot] bourgois [at] protonmail [dot] com