metadata

license: apache-2.0
base_model: sentence-transformers/all-mpnet-base-v2
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: IKT_classifier_netzero_best
    results: []
widget:
  - text: >-
      We have put forth a long-term low- emissions development strategy (LEDS)
      that aspires to halve emissions from its peak to 33 MtCO2e by 2050, with a
      view to achieving net-zero emissions as soon as viable in the second half
      of the century. This will require serious and concerted efforts across our
      industry, economy and society
    example_title: NET-ZERO
  - text: >-
      Unconditional Contribution In the unconditional scenario, GHG emissions
      would be reduced by 27.56 Mt CO2e (6.73%) below BAU in 2030 in the
      respective sectors. 26.3 Mt CO2e (95.4%) of this emission reduction will
      be from the Energy sector while 0.64 (2.3%) and 0.6 (2.2%) Mt CO2e
      reduction will be from AFOLU (agriculture) and waste sector respectively.
      There will be no reduction in the IPPU sector. Conditional Contribution In
      the conditional scenario, GHG emissions would be reduced by 61.9 Mt CO2e
      (15.12%) below BAU in 2030 in the respective sectors.
    example_title: TARGET_FREE
  - text: >-
      This land is buffered from the sea by the dyke and a network of drains and
      pumps will control the water levels in the polder. We have raised the
      minimum platform levels for new developments from 3m to 4m above the
      Singapore Height Datum (SHD) since 2011. Presently, critical
      infrastructure on existing coastal land, notably Changi Airport Terminal 5
      and Tuas Port, will be constructed with platform levels at least 5m above
      SHD.
    example_title: NEGATIVE

IKT_classifier_netzero_best

This model is a fine-tuned version of sentence-transformers/all-mpnet-base-v2 on the GIZ/policy_qa_v0_1 dataset. It achieves the following results on the evaluation set:

Loss: 0.4126
Precision Macro: 0.9246
Precision Weighted: 0.9248
Recall Macro: 0.9209
Recall Weighted: 0.9211
F1-score: 0.9219
Accuracy: 0.9211

Model description

The model is a multi-class text classifier based on sentence-transformers/all-mpnet-base-v2 and fine-tuned on text sourced from national climate policy documents.

Intended uses & limitations

The classifier assigns a class of 'NEGATIVE','TARGET_FREE', or 'NET-ZERO' to denote alignment with Net-Zero targets in extracted passages from the documents. The intended use is for climate policy researchers and analysts seeking to automate the process of reviewing lengthy, non-standardized PDF documents to produce summaries and reports.

The performance of the classifier is very high. On training, the classifier exhibited very good overall performance (F1 ~ 0.9). This performance was evenly balanced between precise identification of true positive classifications (precision ~ 0.9) and a wide net to capture as many true positives as possible (recall ~ 0.9). When tested on real world unseen test data, the performance was still very high (F1 ~ 0.85). However, testing was based on a small out-of-sample dataset. Therefore classification performance will need to further evaluated on deployment.

Training and evaluation data

The training dataset is comprised of labelled passages from 2 sources:

ClimateWatch NDC Sector data.
IKI TraCS Climate Strategies for Transport Tracker implemented by GIZ and funded by theInternational Climate Initiative (IKI) of the German Federal Ministry for Economic Affairs and Climate Action (BMWK). Here we utilized the QA dataset (CW_NDC_data_Sector).

The combined datasetGIZ/policy_qa_v0_1 contains ~85k rows. Each row is duplicated twice, to provide varying sequence lengths (denoted by the values 'small', 'medium', and 'large', which correspond to sequence lengths of 60, 85, and 150 respectively - indicated in the 'strategy' column). This effectively means the dataset is reduced by 1/3 in useful size, and the 'strategy' value should be selected based on the use case. For this training, we utilized the 'medium' samples Furthermore, for each row, the 'context' column contains 3 samples of varying quality. The approach used to assess quality and select samples is described below.

The pre-processing operations used to produce the final training dataset were as follows:

Dataset is filtered based on 'medium' value in 'strategy' column (sequence length = 85).
For ClimateWatch, all rows are removed as there was assessed to be no taxonomical alignment with the IKITracs labels inherent to the dataset. For IKITracs, labels are assigned based on the presence of certain substrings based on 'parameter' values which correspond to assessments of Net-Zero targets by human annotaters. The specific assignments are as follows:
- 'NET-ZERO': target_labels = ['T_Netzero','T_Netzero_C']
- 'NEGATIVE': target_labels_neg = ['T_Economy_C','T_Economy_Unc','T_Adaptation_C','T_Adaptation_Unc','T_Transport_C','T_Transport_O_C','T_Transport_O_Unc','T_Transport_Unc']
- 'TARGET_FREE': random sample of other (non-target) labeled data
If 'context_translated' is available and the 'language' is not English, 'context' is replaced with 'context_translated'.
The dataset is "exploded" - i.e., the text samples in the 'context' column, which are lists, are converted into separate rows - and labels are merged to align with the associated samples.
The 'match_onanswer' and 'answerWordcount' are used conditionally to select high quality samples (prefers high % of word matches in 'match_onanswer', but will take lower if there is a high 'answerWordcount')
Data is then augmented using sentence shuffle from the albumentations library and NLP-based insertions using nlpaug. This is done to increase the number of training samples available for the Net-Zero class from 62 to 124. The end result is a almost equal sample per class breakdown of:
- 'NET-ZERO': 124
- 'NEGATIVE': 126
- 'TARGET_FREE': 125

Training procedure

The model hyperparameters were tuned using optuna over 10 trials on a truncated training and validation dataset. The model was then trained over 5 epochs using the best hyperparameters identified.

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 9.588722322096848e-05
train_batch_size: 3
eval_batch_size: 3
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 400.0
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision Macro	Precision Weighted	Recall Macro	Recall Weighted	F1-score	Accuracy
No log	1.0	113	0.7402	0.8808	0.8847	0.8697	0.8684	0.8694	0.8684
No log	2.0	226	0.8484	0.84	0.8358	0.6752	0.6842	0.6675	0.6842
No log	3.0	339	0.3188	0.9209	0.9229	0.9209	0.9211	0.9200	0.9211
No log	4.0	452	0.5524	0.8889	0.8925	0.8718	0.8684	0.8689	0.8684
0.5553	5.0	565	0.4126	0.9246	0.9248	0.9209	0.9211	0.9219	0.9211

Framework versions

Transformers 4.31.0
Pytorch 2.0.1+cu118
Datasets 2.13.1
Tokenizers 0.13.3