search

tagoyal / sow-reap-paraphrasing

Contains data/code for the paper "Neural Syntactic Preordering for Controlled Paraphrase Generation" (ACL 2020).
77 stars 9 forks source link

how generate train_sow.hdf5 file #3

Closed jiezouguihuafu closed 4 years ago

jiezouguihuafu commented 4 years ago

train_sow.hdf5 I want to train SOW model ，but it need train_sow.hdf5 file . And the file how to get？

tagoyal commented 4 years ago

it's there in the google drive link (in the readme) under the data folder.

jiezouguihuafu commented 4 years ago

it's there in the google drive link (in the readme) under the data folder.

if I want to train at myself data set，then，how to generate this file？

tagoyal commented 4 years ago

Code updated in the preprocessing folder. Let me know if there are issues

TITC commented 3 years ago

when i run the filter_data.py in google colab python filter_data.py --input_file [input_file] --output_file [output_file] --wordEmbed_file [wordEmbed_file] adjust filter_data as below for debug purpose in vscode

if __name__ == '__main__':
    args = parser.parse_args()
    args.input_file = "/content/sow-reap-paraphrasing/sample_test_baseline.txt"
    args.output_file = "/content/sow-reap-paraphrasing/sample_test_output.txt"
    args.wordEmbed_file = "/content/sow-reap-paraphrasing/data/glove/glove.6B.50d.txt"
    input_file = args.input_file
    output_file = args.output_file
    wordEmbed_file = args.wordEmbed_file

and it shows image

when I check the variable embed1, it looks not a numerical vector image