Hugging Face's logo

rafmacalaba
/
gliner_datause

Token Classification
GLiNER
PyTorch
ner
data-use
Model card Files
xet
 Community
rafmacalaba commited on
Commit
45e5283
·
verified ·
1 Parent(s): 28c45da

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +49 -16
README.md CHANGED
@@ -30,25 +30,58 @@ Fine-tune of `urchade/gliner_large-v2.1` for data-use mention extraction
30
 
31
  | thr | tp | fp | fn | precision | recall | f0.5 | f1 |
32
  | --- | --- | --- | --- | --- | --- | --- | --- |
33
- | 0.10 | 16046 | 9301 | 123 | 0.6331 | 0.9924 | 0.6825 | 0.7730 |
34
- | 0.20 | 16016 | 6724 | 153 | 0.7043 | 0.9905 | 0.7475 | 0.8233 |
35
- | 0.30 | 15971 | 5288 | 198 | 0.7513 | 0.9878 | 0.7890 | 0.8534 |
36
- | 0.40 | 15895 | 4053 | 274 | 0.7968 | 0.9831 | 0.8282 | 0.8802 |
37
- | 0.50 | 15694 | 2871 | 475 | 0.8454 | 0.9706 | 0.8678 | 0.9037 |
38
- | 0.60 | 14185 | 1753 | 1984 | 0.8900 | 0.8773 | 0.8874 | 0.8836 |
39
- | 0.70 | 11161 | 827 | 5008 | 0.9310 | 0.6903 | 0.8703 | 0.7928 |
40
 
41
- **Best F0.5**: 0.8874 (thr=0.6)
42
- **Best F1**: 0.9037 (thr=0.5)
43
 
44
  ## Evaluation breakdown (holdout)
45
 
46
  | group | examples | spans | thr | precision | recall | f0.5 | f1 |
47
  | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
48
- | overall | 12531 | 16417 | 0.60 | 0.8900 | 0.8773 | 0.8874 | 0.8836 |
49
- | prwp | 9079 | 12380 | 0.60 | 0.8886 | 0.8935 | 0.8896 | 0.8911 |
50
- | fcv | 3452 | 4037 | 0.50 | 0.8670 | 1.0000 | 0.8907 | 0.9287 |
51
- | general_prwp | 9079 | 12380 | 0.60 | 0.8886 | 0.8935 | 0.8896 | 0.8911 |
52
- | fcv_pads_east_asia | 784 | 863 | 0.50 | 0.8877 | 1.0000 | 0.9081 | 0.9405 |
53
- | jdc_operational | 163 | 178 | 0.60 | 0.9193 | 0.8555 | 0.9058 | 0.8862 |
54
- | refugee_pads | 803 | 875 | 0.50 | 0.8475 | 1.0000 | 0.8742 | 0.9175 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
30
 
31
  | thr | tp | fp | fn | precision | recall | f0.5 | f1 |
32
  | --- | --- | --- | --- | --- | --- | --- | --- |
33
+ | 0.10 | 21384 | 11373 | 306 | 0.6528 | 0.9859 | 0.7001 | 0.7855 |
34
+ | 0.20 | 21251 | 8687 | 439 | 0.7098 | 0.9798 | 0.7512 | 0.8232 |
35
+ | 0.30 | 21069 | 7139 | 621 | 0.7469 | 0.9714 | 0.7831 | 0.8445 |
36
+ | 0.40 | 20815 | 5724 | 875 | 0.7843 | 0.9597 | 0.8141 | 0.8632 |
37
+ | 0.50 | 20317 | 4241 | 1373 | 0.8273 | 0.9367 | 0.8471 | 0.8786 |
38
+ | 0.60 | 19018 | 2836 | 2672 | 0.8702 | 0.8768 | 0.8715 | 0.8735 |
39
+ | 0.70 | 15666 | 1516 | 6024 | 0.9118 | 0.7223 | 0.8663 | 0.8060 |
40
 
41
+ **Best F0.5**: 0.8715 (thr=0.6)
42
+ **Best F1**: 0.8786 (thr=0.5)
43
 
44
  ## Evaluation breakdown (holdout)
45
 
46
  | group | examples | spans | thr | precision | recall | f0.5 | f1 |
47
  | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
48
+ | overall | 18709 | 22052 | 0.60 | 0.8702 | 0.8768 | 0.8715 | 0.8735 |
49
+ | prwp | 9188 | 12359 | 0.60 | 0.8574 | 0.8469 | 0.8553 | 0.8521 |
50
+ | fcv | 9521 | 9693 | 0.60 | 0.8863 | 0.9159 | 0.8921 | 0.9009 |
51
+ | general_prwp | 9188 | 12359 | 0.60 | 0.8574 | 0.8469 | 0.8553 | 0.8521 |
52
+ | fcv_pads_east_africa | 6794 | 6784 | 0.60 | 0.8991 | 0.9156 | 0.9023 | 0.9073 |
53
+ | jdc_operational | 158 | 204 | 0.50 | 0.9212 | 0.9541 | 0.9276 | 0.9373 |
54
+ | refugee_pads | 687 | 743 | 0.70 | 0.9237 | 0.8036 | 0.8969 | 0.8595 |
55
+ | reliefweb | 1882 | 1962 | 0.70 | 0.8920 | 0.7735 | 0.8655 | 0.8286 |
56
+
57
+ ### Per-label (overall)
58
+
59
+ | label | examples | spans | thr | precision | recall | f0.5 | f1 |
60
+ | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
61
+ | NAMED_DATA | 18709 | 8571 | 0.70 | 0.8611 | 0.7311 | 0.8315 | 0.7908 |
62
+ | DESCRIPTIVE_DATA | 18709 | 11348 | 0.70 | 0.8525 | 0.6765 | 0.8103 | 0.7544 |
63
+ | VAGUE_DATA | 18709 | 2133 | 0.60 | 0.6192 | 0.5741 | 0.6096 | 0.5958 |
64
+
65
+ <details>
66
+ <summary>Origin breakdown (per-origin metrics)</summary>
67
+
68
+ | origin | examples | spans | thr | precision | recall | f0.5 | f1 |
69
+ | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
70
+ | fcv_pads_east_africa | 6794 | 6784 | 0.60 | 0.8991 | 0.9156 | 0.9023 | 0.9073 |
71
+ | general_prwp | 9188 | 12359 | 0.60 | 0.8574 | 0.8469 | 0.8553 | 0.8521 |
72
+ | jdc_operational | 158 | 204 | 0.50 | 0.9212 | 0.9541 | 0.9276 | 0.9373 |
73
+ | refugee_pads | 687 | 743 | 0.70 | 0.9237 | 0.8036 | 0.8969 | 0.8595 |
74
+ | reliefweb | 1882 | 1962 | 0.70 | 0.8920 | 0.7735 | 0.8655 | 0.8286 |
75
+
76
+ </details>
77
+
78
+ <details>
79
+ <summary>Per-label details</summary>
80
+
81
+ | label | examples | spans | thr | precision | recall | f0.5 | f1 |
82
+ | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
83
+ | NAMED_DATA | 18709 | 8571 | 0.70 | 0.8611 | 0.7311 | 0.8315 | 0.7908 |
84
+ | DESCRIPTIVE_DATA | 18709 | 11348 | 0.70 | 0.8525 | 0.6765 | 0.8103 | 0.7544 |
85
+ | VAGUE_DATA | 18709 | 2133 | 0.60 | 0.6192 | 0.5741 | 0.6096 | 0.5958 |
86
+
87
+ </details>