search

pkouris / abtextsum

Abstractive text summarization based on deep learning and semantic content generalization
17 stars 4 forks source link

I have a error #1

Open ghost opened 5 years ago

ghost commented 5 years ago

When I run "main.py" , it gives the following error :

FileNotFoundError: [Errno 2] No such file or directory: 'path/to/datasets/directorytrain/train_lg100d5gigaword_article.txt'

Can you help me to solve it ? Thank you very much.

pkouris commented 5 years ago

When I run "main.py" , it gives the following error :

FileNotFoundError: [Errno 2] No such file or directory: 'path/to/datasets/directorytrain/train_lg100d5gigaword_article.txt'

Can you help me to solve it ? Thank you very much.

In order to generalize the dataset according to a model of LG or NEG strategy (as we explain in the paper), you should use the functionality of the file preprocessing.py which produces files such as train_lg100d5gigaword_article.txt. Therefore, the file train_lg100d5gigaword_article.txt should be produced in the phase of text generalization to be used in training after the task of building the dataset.

The file main.py performs the functionality of building dataset, training and testing. The values of hyperparameters, the generalization strategy as well as the paths of the corresponding files should be specified in parameters.py.

Generally, our framework includes the following steps:

  1. Preprocessing of the dataset (preprocessing.py)
  2. Text generalization (preprocessing.py)
  3. Building dataset for training (main.py defining the parameters in parameters.py)
  4. Training (main.py defining the parameters in parameters.py)
  5. Post-processing (postprocessing.py)
  6. Testing

As the source code of this framework includes complicated steps, we will update the readme file specifying these steps, as soon as possible.