Question Answering System using LSTM and BERT

# NQAS ## For Training ``` python train.py --learning_rate 3e-4 --num_train_epochs 5 --per_device_train_batch_size 8 \ --model_name_or_path distilbert-base-uncased --dataset_name squad --output_dir < enter output directory for model saving > \ --do_train --do_eval ``` ## For Inference Model can be found in [google drive](https://drive.google.com/drive/folders/1cXHNbh_ZTOPJTheSmrM6WNAc1RMxbbqw?usp=sharing) edit model path in test.py edit question and context variables in test.py ``` python test.py ``` ## Real-Time Conversation Agent Google CCAI Agent Interface for conversation streaming with an agent in real-time. You can find the agent as ``` agent.json ``` with some examples.