ai-club-inductions-21-nlp-ELECTRA-base-squad

This model is the deepset/electra-base-squad2 pre-trained model trained on data from AI Inductions 21 NLP competition (https://www.kaggle.com/c/ai-inductions-21-nlp) for extractive QA.

Model description

More information needed

Intended uses & limitations

AI Inductions 21 NLP competition

Training and evaluation data

AI Inductions 21 NLP competition data

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• max_length = 512
• doc_stride = 384
• learning_rate: 2e-05
• weight_decay=0.01
• 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: 10

Framework versions

• Transformers 4.11.3
• Pytorch 1.7.1+cpu
• Datasets 1.14.0
• Tokenizers 0.10.3