Hugging Face's logo

Dagobert42
/
distilbert-base-uncased-biored-finetuned

Token Classification
Transformers
Safetensors
English
distilbert
low-resource NER
token_classification
biomedicine
medical NER
Generated from Trainer
Inference Endpoints
Model card Files Community
Dagobert42 commited on
Commit
d61c25d
•
1 Parent(s): fd549c6

Push distilbert-base-uncased trained on biored-original_splits.pt (800 samples)

Browse files
Files changed (3) hide show
  1. README.md +14 -20
  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 [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the bigbio/biored dataset.
30
  It achieves the following results on the evaluation set:
31
- - Loss: 0.7959
32
- - Accuracy: 0.752
33
- - Precision: 0.5065
34
- - Recall: 0.3768
35
- - F1: 0.4166
36
- - Weighted F1: 0.7248
37
 
38
  ## Model description
39
 
@@ -64,20 +64,14 @@ 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 | 13 | 1.0377 | 0.7028 | 0.1687 | 0.1581 | 0.1469 | 0.6075 |
68
- | No log | 2.0 | 26 | 0.9390 | 0.7165 | 0.4311 | 0.1589 | 0.1477 | 0.6128 |
69
- | No log | 3.0 | 39 | 0.9029 | 0.7229 | 0.5278 | 0.2131 | 0.2335 | 0.6336 |
70
- | No log | 4.0 | 52 | 0.8697 | 0.7279 | 0.4497 | 0.2345 | 0.2596 | 0.642 |
71
- | No log | 5.0 | 65 | 0.8714 | 0.7305 | 0.4944 | 0.252 | 0.2772 | 0.6437 |
72
- | No log | 6.0 | 78 | 0.8372 | 0.7356 | 0.5031 | 0.2899 | 0.3201 | 0.6679 |
73
- | No log | 7.0 | 91 | 0.8309 | 0.7387 | 0.5018 | 0.3033 | 0.3431 | 0.6776 |
74
- | No log | 8.0 | 104 | 0.8381 | 0.7396 | 0.5106 | 0.3116 | 0.3458 | 0.6821 |
75
- | No log | 9.0 | 117 | 0.8181 | 0.7419 | 0.4883 | 0.3353 | 0.3759 | 0.6965 |
76
- | No log | 10.0 | 130 | 0.8079 | 0.7427 | 0.4991 | 0.3462 | 0.3875 | 0.7029 |
77
- | No log | 11.0 | 143 | 0.8272 | 0.7434 | 0.5075 | 0.3578 | 0.3896 | 0.7027 |
78
- | No log | 12.0 | 156 | 0.8047 | 0.7449 | 0.5 | 0.3772 | 0.4145 | 0.714 |
79
- | No log | 13.0 | 169 | 0.8272 | 0.7498 | 0.5196 | 0.3678 | 0.4078 | 0.7107 |
80
- | No log | 14.0 | 182 | 0.8255 | 0.7498 | 0.515 | 0.3819 | 0.4217 | 0.7161 |
81
 
82
 
83
  ### Framework versions
 
28
 
29
  This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the bigbio/biored dataset.
30
  It achieves the following results on the evaluation set:
31
+ - Loss: 0.5577
32
+ - Accuracy: 0.8101
33
+ - Precision: 0.605
34
+ - Recall: 0.5358
35
+ - F1: 0.5617
36
+ - Weighted F1: 0.8043
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 | 100 | 0.7731 | 0.7459 | 0.5685 | 0.3108 | 0.3467 | 0.6748 |
68
+ | No log | 2.0 | 200 | 0.6716 | 0.7705 | 0.6073 | 0.4069 | 0.4512 | 0.7384 |
69
+ | No log | 3.0 | 300 | 0.6072 | 0.7922 | 0.6097 | 0.4576 | 0.5126 | 0.7661 |
70
+ | No log | 4.0 | 400 | 0.6077 | 0.7965 | 0.6042 | 0.4866 | 0.5285 | 0.7732 |
71
+ | 0.648 | 5.0 | 500 | 0.5842 | 0.8006 | 0.5972 | 0.5243 | 0.5518 | 0.7937 |
72
+ | 0.648 | 6.0 | 600 | 0.5688 | 0.8068 | 0.5993 | 0.534 | 0.5611 | 0.8004 |
73
+ | 0.648 | 7.0 | 700 | 0.5869 | 0.8061 | 0.6039 | 0.5313 | 0.5569 | 0.7958 |
74
+ | 0.648 | 8.0 | 800 | 0.5740 | 0.8133 | 0.6117 | 0.5324 | 0.5658 | 0.8025 |
 
 
 
 
 
 
75
 
76
 
77
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cdd6ed4b3227bc3fd1a3216ec56c95758d95a28ddf62ca682cb4798c592d8762
3
  size 265485396
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f34561e38eb34da23ade0d8b1b536275154f1323d3f21af691e0280d9bbdf822
3
  size 265485396
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6f6f2ce0f4d9c1ba8d683f81ffab86087a2bf5dc5986cbef865b108a7f29692f
3
  size 4219
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ec3cb3dca4214eca559c111a93e2ffa6cfe624987e026ff872f4ed30ff555cb
3
  size 4219