|
--- |
|
license: apache-2.0 |
|
base_model: alex-miller/ODABert |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: iati-drr-classifier |
|
results: [] |
|
datasets: |
|
- alex-miller/iati-policy-markers |
|
pipeline_tag: text-classification |
|
widget: |
|
- text: "Core pandemic preparedness and response and integrated global health priorities" |
|
example_title: "Positive" |
|
- text: "Education programs for the disabled and access to learning opportunities for children" |
|
example_title: "Negative" |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# iati-drr-classifier |
|
|
|
This model is a fine-tuned version of [alex-miller/ODABert](https://huggingface.co/alex-miller/ODABert) on a subset of the [alex-miller/iati-policy-markers](https://huggingface.co/datasets/alex-miller/iati-policy-markers) dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3910 |
|
- Accuracy: 0.8207 |
|
|
|
## Model description |
|
|
|
This model has been trained to identify disaster risk reduction (DRR) project titles and/or descriptions. It returns "0" for projects with no DRR component, and "1" for projects with DRR as a principal or significant objective. |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
More information needed |
|
|
|
## Training procedure |
|
|
|
Code to subset the dataset and train the model is available [here](https://github.com/akmiller01/iati-policy-marker-hf-dataset/blob/main/use_cases/drr_train.ipynb). |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 1e-06 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 4 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 0.607 | 1.0 | 508 | 0.5053 | 0.7507 | |
|
| 0.4569 | 2.0 | 1016 | 0.4289 | 0.7980 | |
|
| 0.4011 | 3.0 | 1524 | 0.4009 | 0.8143 | |
|
| 0.3786 | 4.0 | 2032 | 0.3910 | 0.8207 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
