Hugging Face's logo

Dagobert42
/
mobilebert-uncased-biored-finetuned

Token Classification
Transformers
Safetensors
English
mobilebert
low-resource NER
token_classification
biomedicine
medical NER
Generated from Trainer
Inference Endpoints
Model card Files Community
Dagobert42 commited on
Commit
6d4e0a4
•
1 Parent(s): 14184ac

Push google/mobilebert-uncased trained on biored-original_splits.pt

Browse files
Files changed (3) hide show
  1. README.md +23 -21
  2. model.safetensors +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -28,12 +28,12 @@ should probably proofread and complete it, then remove this comment. -->
28
 
29
  This model is a fine-tuned version of [mobilebert-uncased](https://huggingface.co/mobilebert-uncased) on the bigbio/biored dataset.
30
  It achieves the following results on the evaluation set:
31
- - Loss: 0.7686
32
- - Accuracy: 0.7387
33
- - Precision: 0.2041
34
- - Recall: 0.2219
35
- - F1: 0.1908
36
- - Weighted F1: 0.683
37
 
38
  ## Model description
39
 
@@ -64,22 +64,24 @@ The following hyperparameters were used during training:
64
 
65
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Weighted F1 |
66
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:-----------:|
67
- | No log | 1.0 | 25 | 1.2311 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
68
- | No log | 2.0 | 50 | 1.0356 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
69
  | No log | 3.0 | 75 | 1.0300 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
70
- | No log | 4.0 | 100 | 1.0246 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
71
- | No log | 5.0 | 125 | 1.0162 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
72
- | No log | 6.0 | 150 | 1.0039 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
73
- | No log | 7.0 | 175 | 0.9806 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
74
- | No log | 8.0 | 200 | 0.9148 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
75
- | No log | 9.0 | 225 | 0.8715 | 0.7187 | 0.2116 | 0.1604 | 0.1484 | 0.6172 |
76
- | No log | 10.0 | 250 | 0.8303 | 0.7261 | 0.1555 | 0.1972 | 0.1737 | 0.6508 |
77
- | No log | 11.0 | 275 | 0.8216 | 0.7292 | 0.1572 | 0.2018 | 0.1764 | 0.6554 |
78
- | No log | 12.0 | 300 | 0.8044 | 0.7299 | 0.2295 | 0.2081 | 0.1786 | 0.6605 |
79
- | No log | 13.0 | 325 | 0.8108 | 0.732 | 0.2304 | 0.2091 | 0.1797 | 0.662 |
80
- | No log | 14.0 | 350 | 0.7920 | 0.7306 | 0.2062 | 0.22 | 0.1877 | 0.6711 |
81
- | No log | 15.0 | 375 | 0.8025 | 0.7332 | 0.2164 | 0.2153 | 0.1836 | 0.6674 |
82
- | No log | 16.0 | 400 | 0.7937 | 0.7335 | 0.1982 | 0.2248 | 0.2039 | 0.6813 |
 
 
83
 
84
 
85
  ### Framework versions
 
28
 
29
  This model is a fine-tuned version of [mobilebert-uncased](https://huggingface.co/mobilebert-uncased) on the bigbio/biored dataset.
30
  It achieves the following results on the evaluation set:
31
+ - Loss: 0.7632
32
+ - Accuracy: 0.7385
33
+ - Precision: 0.2012
34
+ - Recall: 0.2384
35
+ - F1: 0.215
36
+ - Weighted F1: 0.7009
37
 
38
  ## Model description
39
 
 
64
 
65
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Weighted F1 |
66
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:-----------:|
67
+ | No log | 1.0 | 25 | 1.2345 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
68
+ | No log | 2.0 | 50 | 1.0379 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
69
  | No log | 3.0 | 75 | 1.0300 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
70
+ | No log | 4.0 | 100 | 1.0228 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
71
+ | No log | 5.0 | 125 | 1.0144 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
72
+ | No log | 6.0 | 150 | 0.9994 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
73
+ | No log | 7.0 | 175 | 0.9681 | 0.7114 | 0.1016 | 0.1429 | 0.1188 | 0.5914 |
74
+ | No log | 8.0 | 200 | 0.8869 | 0.7147 | 0.2167 | 0.1487 | 0.1303 | 0.6007 |
75
+ | No log | 9.0 | 225 | 0.8511 | 0.7242 | 0.2064 | 0.1716 | 0.1598 | 0.6298 |
76
+ | No log | 10.0 | 250 | 0.8187 | 0.7287 | 0.157 | 0.1991 | 0.1754 | 0.653 |
77
+ | No log | 11.0 | 275 | 0.8046 | 0.7317 | 0.1581 | 0.2035 | 0.1775 | 0.6581 |
78
+ | No log | 12.0 | 300 | 0.7900 | 0.732 | 0.1935 | 0.2126 | 0.1887 | 0.6688 |
79
+ | No log | 13.0 | 325 | 0.7865 | 0.734 | 0.2312 | 0.2129 | 0.1828 | 0.6664 |
80
+ | No log | 14.0 | 350 | 0.7758 | 0.7346 | 0.1604 | 0.2148 | 0.1819 | 0.6672 |
81
+ | No log | 15.0 | 375 | 0.7958 | 0.7376 | 0.2086 | 0.2141 | 0.1884 | 0.6697 |
82
+ | No log | 16.0 | 400 | 0.7757 | 0.733 | 0.2002 | 0.2347 | 0.2122 | 0.6904 |
83
+ | No log | 17.0 | 425 | 0.7874 | 0.7393 | 0.2067 | 0.2196 | 0.2119 | 0.6828 |
84
+ | No log | 18.0 | 450 | 0.7915 | 0.735 | 0.2043 | 0.2391 | 0.2197 | 0.6959 |
85
 
86
 
87
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b96fb4143c4d20903757eeceec5b505268946dbb173fdfb2b520ad4ecde10729
3
  size 98480380
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd5480e760f36c5e3e61c261707a26d1a9724d097c64735602cebf3154ab7f25
3
  size 98480380
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8c46b669772226cd7bce858b7531b3b0ac2681805b2bd702e5228fed1a203250
3
  size 4219
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b09ada94fdc9017cf52accd1cdd1f264a542794b0fa5169933a874d0b964f6df
3
  size 4219