jialinselenasong
commited on
Commit
•
b2d3a67
1
Parent(s):
00b5950
Training complete
Browse files
README.md
ADDED
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: NeuML/pubmedbert-base-embeddings
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
metrics:
|
7 |
+
- precision
|
8 |
+
- recall
|
9 |
+
- f1
|
10 |
+
- accuracy
|
11 |
+
model-index:
|
12 |
+
- name: pubmed-bert-all-deep
|
13 |
+
results: []
|
14 |
+
---
|
15 |
+
|
16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
17 |
+
should probably proofread and complete it, then remove this comment. -->
|
18 |
+
|
19 |
+
# pubmed-bert-all-deep
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [NeuML/pubmedbert-base-embeddings](https://huggingface.co/NeuML/pubmedbert-base-embeddings) on an unknown dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 0.9764
|
24 |
+
- Precision: 0.4738
|
25 |
+
- Recall: 0.4800
|
26 |
+
- F1: 0.4769
|
27 |
+
- Accuracy: 0.7380
|
28 |
+
|
29 |
+
## Model description
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Intended uses & limitations
|
34 |
+
|
35 |
+
More information needed
|
36 |
+
|
37 |
+
## Training and evaluation data
|
38 |
+
|
39 |
+
More information needed
|
40 |
+
|
41 |
+
## Training procedure
|
42 |
+
|
43 |
+
### Training hyperparameters
|
44 |
+
|
45 |
+
The following hyperparameters were used during training:
|
46 |
+
- learning_rate: 2e-05
|
47 |
+
- train_batch_size: 8
|
48 |
+
- eval_batch_size: 8
|
49 |
+
- seed: 42
|
50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
51 |
+
- lr_scheduler_type: linear
|
52 |
+
- num_epochs: 10
|
53 |
+
|
54 |
+
### Training results
|
55 |
+
|
56 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
57 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
58 |
+
| No log | 1.0 | 363 | 1.1611 | 0.1988 | 0.1654 | 0.1806 | 0.6258 |
|
59 |
+
| 1.3011 | 2.0 | 726 | 1.0030 | 0.3355 | 0.3221 | 0.3287 | 0.6877 |
|
60 |
+
| 0.9032 | 3.0 | 1089 | 0.9300 | 0.4125 | 0.3563 | 0.3823 | 0.7095 |
|
61 |
+
| 0.9032 | 4.0 | 1452 | 0.8892 | 0.4466 | 0.4189 | 0.4323 | 0.7220 |
|
62 |
+
| 0.7036 | 5.0 | 1815 | 0.9079 | 0.4476 | 0.4530 | 0.4503 | 0.7257 |
|
63 |
+
| 0.5735 | 6.0 | 2178 | 0.9415 | 0.4651 | 0.4684 | 0.4667 | 0.7299 |
|
64 |
+
| 0.4796 | 7.0 | 2541 | 0.9484 | 0.4791 | 0.4558 | 0.4672 | 0.7324 |
|
65 |
+
| 0.4796 | 8.0 | 2904 | 0.9677 | 0.4673 | 0.4757 | 0.4715 | 0.7335 |
|
66 |
+
| 0.4197 | 9.0 | 3267 | 0.9810 | 0.4760 | 0.4791 | 0.4775 | 0.7361 |
|
67 |
+
| 0.3812 | 10.0 | 3630 | 0.9764 | 0.4738 | 0.4800 | 0.4769 | 0.7380 |
|
68 |
+
|
69 |
+
|
70 |
+
### Framework versions
|
71 |
+
|
72 |
+
- Transformers 4.40.1
|
73 |
+
- Pytorch 2.2.1+cu121
|
74 |
+
- Datasets 2.19.1
|
75 |
+
- Tokenizers 0.19.1
|