Codes in this repo are based on the paper Filling the Gap of Utterance-aware and Speaker-aware Representation for Multi-turn Dialogue and the accompanying github https://github.com/comprehensiveMap/MDFN

Instruction

Our code is compatible with compatible with python 3.x so for all commands listed below python is python3.

We strongly suggest you to use conda to control the virtual environment.

• Install requirements

pip install -r requirements.txt

• To run the Electra Basline

python run_MDFN.py \
--data_dir datasets/mutual \
--model_name_or_path \
google/electra-large-discriminator \
--model_type electra \
--baseline \
--task_name mutual\
--output_dir output_mutual_electra \
--cache_dir cached_models \

• To run the MDFN Baseline

python run_MDFN.py \
--data_dir datasets/mutual \
--model_name_or_path \
google/electra-large-discriminator \
--model_type electra \
--task_name mutual\
--output_dir output_mutual_electra \
--cache_dir cached_models \

• To run the Speaker-Aware Embedding

python main.py \
--baseline \
--speaker_aware \
--data_dir datasets/mutual \
--model_name_or_path google/electra-large-discriminator \
--model_type electra \
--task_name mutual \
--output_dir output_mutual_electra \
--cache_dir cached_models \
--max_seq_length 256 \
--do_train --do_eval \
--train_batch_size 2 --eval_batch_size 2 \
--learning_rate 4e-6 --num_train_epochs 6 \
--gradient_accumulation_steps 1 --local_rank -1

• To run the Speaker-Aware Decouple

python main.py \
--data_dir datasets/mutual \
--model_name_or_path google/electra-large-discriminator \
--model_type electra \
--task_name mutual \
--output_dir output_mutual_electra \
--cache_dir cached_models \
--max_seq_length 256 \
--do_train --do_eval \
--train_batch_size 2 --eval_batch_size 2 \
--learning_rate 4e-6 --num_train_epochs 6 \
--gradient_accumulation_steps 1 --local_rank -1

• To run the Response-Aware Query

python main.py \
--response_aware \
--data_dir datasets/mutual \
--model_name_or_path google/electra-large-discriminator \
--model_type electra \
--task_name mutual \
--output_dir output_mutual_electra \
--cache_dir cached_models \
--max_seq_length 256 \
--do_train --do_eval \
--train_batch_size 2 --eval_batch_size 2 \
--learning_rate 4e-6 --num_train_epochs 6 \
--gradient_accumulation_steps 1 --local_rank -1

• To run the Response-Aware BiDAF

python main.py \
--BiDAF \
--data_dir datasets/mutual \
--model_name_or_path google/electra-large-discriminator \
--model_type electra \
--task_name mutual \
--output_dir output_mutual_electra \
--cache_dir cached_models \
--max_seq_length 256 \
--do_train --do_eval \
--train_batch_size 2 --eval_batch_size 2 \
--learning_rate 4e-6 --num_train_epochs 6 \
--gradient_accumulation_steps 1 --local_rank -1

• To run the Topic-Aware Representation

python3 run_ELECTRA_topic.py
--do_topic 
--data_dir datasets/mutual 
--model_name_or_path google/electra-large-discriminator 
--model_type electra 
--task_name mutual 
--output_dir output_mutual_topic_att 
--cache_dir cached_models 
--max_seq_length 256 
--do_train --do_eval 
--train_batch_size 4 --eval_batch_size 4 
--learning_rate 4e-6 --num_train_epochs 3 
--gradient_accumulation_steps 1 --local_rank -1