aap9002/NLI-BILSTM

language: en license: cc-by-4.0 tags:

• text-classification repo: https://github.com/AAP9002/COMP34812-NLU-NLI

---

Model Card for z72819ap-e91802zc-NLI

This is a classification model that was trained to detect whether a premise and hypothesis entail each other or not, using binary classification.

Model Details

Model Description

This model is based upon the Enhanced LSTM for Natural Language Inference architecture using BILSTM instead of LSTM trained on over 24K premise-hypothesis pairs from the shared task dataset for Natural Language Inference (NLI).

• Developed by: Alan Prophett and Zac Curtis
• Language(s): English
• Model type: Supervised
• Model architecture: BILSTM
• Finetuned from model [optional]: None

Model Resources

• Repository: None
• Paper or documentation: None

Training Details

Training Data

24K+ premise-hypothesis pairs from the shared task dataset provided for Natural Language Inference (NLI).

Training Procedure

Training Hyperparameters

  - seed: 42
  - learning_rate: 1e-04
  - train_batch_size: 64
  - eval_batch_size: 64
  - num_epochs: 20
  

Speeds, Sizes, Times

  - overall training time: 3  minutes 4 seconds
  - duration per training epoch: 34 seconds
  - model size: 30.7 MB

Evaluation

Testing Data & Metrics

Testing Data

A subset of the development set provided, amounting to 6K+ pairs.

Metrics

  - Recall
  - F1-score
  - Accuracy

Results

The BILSTM RNN Model obtained an F1-score of 70% and an accuracy of 70%.

Technical Specifications

Hardware

  - RAM: at least 25 GB
  - Storage: at least 38.1 GB,
  - GPU: a100 40GB

Software

  - Tensorflow 2.18.0+cu12.4
  - Pandas 2.2.2
  - NumPy 2.0.2
  - Seaborn 0.13.2
  - Matplotlib 3.10.0
  - Scikit-learn 1.6.1

Bias, Risks, and Limitations

Any inputs (concatenation of two sequences) longer than 512 subwords will be truncated by the model.

Additional Information

The hyperparameters were determined by experimentation with different values.