File size: 2,386 Bytes
8c44129
 
 
 
bbffacc
8c44129
bbffacc
8c44129
bbffacc
 
 
 
 
 
 
6aed013
 
 
 
 
 
 
 
 
8c44129
 
 
 
 
bed5460
bbffacc
e88d51c
bbffacc
22900d9
8c44129
22900d9
8c44129
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bbffacc
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
---
license: apache-2.0
tags:
- generated_from_trainer
- medical
model-index:
- name: stop_reasons_classificator_multilabel
  results: []
datasets:
- opentargets/clinical_trial_reason_to_stop
language:
- en
metrics:
- accuracy
library_name: transformers
widget:
  - text: "Study stopped due to problems to recruit patients"
    example_title: "Enrollment issues"
  - text: "Efficacy endpoint unmet"
    example_title: "Negative reasons"
  - text: "Study stopped due to unexpected adverse effects"
    example_title: "Safety"
  - text: "Study paused due to the pandemic"
    example_title: "COVID-19"
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# Clinical trial stop reasons

This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the task of classification of why a clinical trial has stopped early.

The dataset containing 3,747 manually curated reasons used for fine-tuning is available in the [Hub](https://huggingface.co/datasets/opentargets/clinical_trial_reason_to_stop).

More details on the model training are available in the GitHub project ([link](https://github.com/opentargets/stopReasons)) and in the associated publication (TBC).


## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 7

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy Thresh |
|:-------------:|:-----:|:----:|:---------------:|:---------------:|
| No log        | 1.0   | 106  | 0.1824          | 0.9475          |
| No log        | 2.0   | 212  | 0.1339          | 0.9630          |
| No log        | 3.0   | 318  | 0.1109          | 0.9689          |
| No log        | 4.0   | 424  | 0.0988          | 0.9741          |
| 0.1439        | 5.0   | 530  | 0.0943          | 0.9743          |
| 0.1439        | 6.0   | 636  | 0.0891          | 0.9763          |
| 0.1439        | 7.0   | 742  | 0.0899          | 0.9760          |


### Framework versions

- Transformers 4.26.0
- Pytorch 1.12.1+cu102
- Datasets 2.9.0
- Tokenizers 0.13.2