Hugging Face's logo

pabRomero
/
BioMedRoBERTa-full-finetuned-ner-pablo

Token Classification
Transformers
TensorBoard
ONNX
Safetensors
roberta
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
pabRomero commited on
Commit
a57b3c9
·
verified ·
1 Parent(s): 775fe43

Training complete

Browse files
Files changed (2) hide show
  1. README.md +15 -20
  2. runs/Sep05_12-11-55_83295d15965e/events.out.tfevents.1725538316.83295d15965e.5325.5 +2 -2
README.md CHANGED
@@ -20,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [allenai/biomed_roberta_base](https://huggingface.co/allenai/biomed_roberta_base) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.0951
24
- - Precision: 0.8139
25
- - Recall: 0.8085
26
- - F1: 0.8112
27
- - Accuracy: 0.9769
28
 
29
  ## Model description
30
 
@@ -43,32 +43,27 @@ More information needed
43
  ### Training hyperparameters
44
 
45
  The following hyperparameters were used during training:
46
- - learning_rate: 0.0002
47
- - train_batch_size: 16
48
- - eval_batch_size: 16
49
  - seed: 42
50
- - gradient_accumulation_steps: 4
51
- - total_train_batch_size: 64
52
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
53
  - lr_scheduler_type: linear
54
- - lr_scheduler_warmup_ratio: 0.05
55
- - num_epochs: 5
56
  - mixed_precision_training: Native AMP
57
 
58
  ### Training results
59
 
60
- | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
61
- |:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
62
- | No log | 0.9970 | 252 | 0.0910 | 0.7348 | 0.7922 | 0.7624 | 0.9723 |
63
- | 0.19 | 1.9980 | 505 | 0.0808 | 0.8030 | 0.7908 | 0.7969 | 0.9768 |
64
- | 0.19 | 2.9990 | 758 | 0.0802 | 0.8125 | 0.7932 | 0.8028 | 0.9768 |
65
- | 0.0433 | 4.0 | 1011 | 0.0865 | 0.8131 | 0.8103 | 0.8117 | 0.9772 |
66
- | 0.0433 | 4.9852 | 1260 | 0.0951 | 0.8139 | 0.8085 | 0.8112 | 0.9769 |
67
 
68
 
69
  ### Framework versions
70
 
71
  - Transformers 4.44.2
72
- - Pytorch 2.4.0+cu121
73
  - Datasets 2.21.0
74
  - Tokenizers 0.19.1
 
20
 
21
  This model is a fine-tuned version of [allenai/biomed_roberta_base](https://huggingface.co/allenai/biomed_roberta_base) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.0596
24
+ - Precision: 0.8360
25
+ - Recall: 0.8339
26
+ - F1: 0.8350
27
+ - Accuracy: 0.9815
28
 
29
  ## Model description
30
 
 
43
  ### Training hyperparameters
44
 
45
  The following hyperparameters were used during training:
46
+ - learning_rate: 5e-05
47
+ - train_batch_size: 32
48
+ - eval_batch_size: 32
49
  - seed: 42
 
 
50
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
  - lr_scheduler_type: linear
52
+ - lr_scheduler_warmup_ratio: 0.1
53
+ - num_epochs: 2
54
  - mixed_precision_training: Native AMP
55
 
56
  ### Training results
57
 
58
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
59
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
60
+ | No log | 1.0 | 231 | 0.0783 | 0.7809 | 0.8068 | 0.7936 | 0.9766 |
61
+ | No log | 2.0 | 462 | 0.0596 | 0.8360 | 0.8339 | 0.8350 | 0.9815 |
 
 
 
62
 
63
 
64
  ### Framework versions
65
 
66
  - Transformers 4.44.2
67
+ - Pytorch 2.4.1+cu121
68
  - Datasets 2.21.0
69
  - Tokenizers 0.19.1
runs/Sep05_12-11-55_83295d15965e/events.out.tfevents.1725538316.83295d15965e.5325.5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:87c63f336006afdb99b438233af61b7b47d1c145a7e6d2723d477556a5b3fcbb
3
- size 6205
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a20e2a4f7183dc37f8f35e9c361b585c5cef4e09d266514f0aafb3ba2c6dedd
3
+ size 7031