---
license: mit
base_model: BAAI/bge-base-en-v1.5
tags:
- generated_from_trainer
model-index:
- name: TAPP-multilabel-bge
  results: []
datasets:
- GIZ/policy_classification
co2_eq_emissions:
  emissions: 71.4552917731392
  source: codecarbon
  training_type: fine-tuning
  on_cloud: true
  cpu_model: Intel(R) Xeon(R) CPU @ 2.30GHz
  ram_total_size: 12.6747894287109
  hours_used: 1.36
  hardware_used: 1 x Tesla T4
---

<!-- 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. -->

# TAPP-multilabel-bge

This model is a fine-tuned version of [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5) on the [Policy-Classification](https://huggingface.co/datasets/GIZ/policy_classification) dataset.

*The loss function BCEWithLogitsLoss is modified with pos_weight to focus on recall, therefore instead of loss the evaluation metrics are used to assess the model performance during training*
It achieves the following results on the evaluation set:

- Precision-micro: 0.7772
- Precision-samples: 0.7644
- Precision-weighted: 0.7756
- Recall-micro: 0.8329
- Recall-samples: 0.7920
- Recall-weighted: 0.8329
- F1-micro: 0.8041
- F1-samples: 0.7609
- F1-weighted: 0.8029

## Model description

The purpose of this model is to predict multiple labels simultaneously from a given input data. Specifically, the model will predict four labels - 
ActionLabel, PlansLabel, PolicyLabel, and TargetLabel - that are relevant to a particular task or application
- **Target**: Targets are an intention to achieve a specific result, for example, to reduce GHG emissions to a specific level 
            (a GHG target) or increase energy efficiency or renewable energy to a specific level (a non-GHG target), typically by  
            a certain date.
- **Action**: Actions are an intention to implement specific means of achieving GHG reductions, usually in forms of concrete projects.
- **Policies**: Policies are domestic planning documents such as policies, regulations or guidlines.
- **Plans**:Plans  are broader than specific policies or actions, such as a general intention to ‘improve efficiency’, ‘develop renewable energy’, etc.

*The terms come from the World Bank's NDC platform and WRI's publication*

## Intended uses & limitations

More information needed

## Training and evaluation data

- Training Dataset: 10031
| Class | Positive Count of Class|
|:-------------|:--------|
| Action | 5416 |
| Plans | 2140 |
| Policy | 1396|
| Target | 2911 |

- Validation Dataset: 932
| Class | Positive Count of Class|
|:-------------|:--------|
| Action | 513 |
| Plans | 198 |
| Policy | 122 |
| Target | 256 |

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 7.4e-05
- 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: cosine
- lr_scheduler_warmup_steps: 200
- num_epochs: 7

### Training results

| Training Loss | Epoch | Step | Validation Loss | Precision-micro | Precision-samples | Precision-weighted | Recall-micro | Recall-samples | Recall-weighted | F1-micro | F1-samples | F1-weighted |
|:-------------:|:-----:|:----:|:---------------:|:---------------:|:-----------------:|:------------------:|:------------:|:--------------:|:---------------:|:--------:|:----------:|:-----------:|
| 0.7161        | 1.0   | 627  | 0.6322          | 0.5931          | 0.6373            | 0.6274             | 0.8219       | 0.7833         | 0.8219          | 0.6890   | 0.6728     | 0.7000      |
| 0.4549        | 2.0   | 1254 | 0.5420          | 0.6639          | 0.6891            | 0.7049             | 0.8090       | 0.7684         | 0.8090          | 0.7293   | 0.7048     | 0.7409      |
| 0.2599        | 3.0   | 1881 | 0.6966          | 0.7354          | 0.7396            | 0.7346             | 0.8219       | 0.7845         | 0.8219          | 0.7762   | 0.7425     | 0.7713      |
| 0.1405        | 4.0   | 2508 | 0.7530          | 0.7569          | 0.7494            | 0.7569             | 0.8292       | 0.7899         | 0.8292          | 0.7914   | 0.7505     | 0.7905      |
| 0.0681        | 5.0   | 3135 | 0.8234          | 0.7596          | 0.7535            | 0.7599             | 0.8356       | 0.7945         | 0.8356          | 0.7958   | 0.7546     | 0.7953      |
| 0.0291        | 6.0   | 3762 | 0.8849          | 0.7773          | 0.7640            | 0.7776             | 0.8301       | 0.7890         | 0.8301          | 0.8028   | 0.7597     | 0.8027      |
| 0.0147        | 7.0   | 4389 | 0.9217          | 0.7772          | 0.7644            | 0.7756             | 0.8329       | 0.7920         | 0.8329          | 0.8041   | 0.7609     | 0.8029      |

|label          | precision |recall |f1-score| support|
|:-------------:|:---------:|:-----:|:------:|:------:|
|Action	|0.826   	|0.883  |0.853   |	513.0  |
|Plans	        |0.653	    |0.646  |0.649   |	198.0  |
|Policy	|0.726      |0.803  |0.762   |	122.0  |
|Target	    |0.791      |0.890  |0.838   |	256.0  |

### Environmental Impact
Carbon emissions were measured using [CodeCarbon](https://github.com/mlco2/codecarbon).
- **Carbon Emitted**: 0.07145 kg of CO2
- **Hours Used**: 1.36 hours

### Training Hardware
- **On Cloud**: yes
- **GPU Model**: 1 x Tesla T4
- **CPU Model**: Intel(R) Xeon(R) CPU @ 2.30GHz
- **RAM Size**: 12.67 GB

### Framework versions

- Transformers 4.38.1
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2