Method for Loading Weights and Saving Them

SeanNaren / deepspeech.torch

Speech Recognition using DeepSpeech2 network and the CTC activation function.

MIT License

259 stars 73 forks source link

Method for Loading Weights and Saving Them #64

Closed nn-learner closed 7 years ago

nn-learner commented 8 years ago

Hi Sean,

I was wondering if there was a way to load the weights of the model, train it on another dataset, and save the weights again. Currently, you are only able to either load or save the weights.

SeanNaren commented 8 years ago

Hey on the main branch you should be able to load (and save) a model after training. You do need to setup the dataset using the wiki guide here. You can load a model like this:

th Train.lua -loadModel -modelPath Path/To/Model.t7.

Hopefully this helps!