Skip to content

can not reproduce the model #8

@ForestLee

Description

@ForestLee

I tested with your pretrained model(neural_char.h5), when i input 'from', i got below:

{"data":{"results":["from keras.layers.core import Dense, Activation","from keras.layers.recurrent import Sequential","from keras.layers.core import Dense"]}}

And i run scrape_data.py, then train model, when i input 'from', my model responsed as below, it was not a sentence but some files content.
1

I opened models with Netron, and found your pre-trained model was different with mine. (Left is yours and right are mine)
2

Do you know any problem with that ? Could you tell me how to reproduce your model?

Thanks

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions