can I use this code to generate a summary from a new source text

[hanyegirl]

i want to generate a summary ,and the source text is not in the text.bin ,just an ordinary English text,what can i do for this task ,thank you for your any help

[iwangjian]

The processing from text data to .bin file is referenced by https://github.com/abisee/cnn-dailymail, you can refer to this code to process your own data.

[hanyegirl]

The processing from text data to .bin file is referenced by https://github.com/abisee/cnn-dailymail, you can refer to this code to process your own data. then use this python3 main.py --mode=decode --data_path=./data/test.bin --vocab_path=./data/vocab --log_root=./log --single_pass=True to generate the relative summary?

[iwangjian]

Yes, but firstly you need to train a model using your own data following the instructions in README.md, and then use the trained model to generate the relative summary.

[hanyegirl]

thank you ,your advice is very important to me

[HunterYCG]

May I ask, after you use the new data set to train the model, what is the test result? At present, I am also doing some related work, so I would like to consult you.

[iwangjian]

If you use a new dataset to train the model, the test input are texts (or articles) in your test set, the test output are corresponding generated summaries.

[HunterYCG]

Thank you very much for your reply.