Handle training sentences that are longer than max_sequence_length

senarvi / theanolm

TheanoLM is a recurrent neural network language modeling tool implemented using Theano

Apache License 2.0

81 stars 29 forks source link

Closed senarvi closed 8 years ago

senarvi commented 8 years ago

Split training sentences that are longer than max_sequence_length and create multiple sequences for a single sentence.

senarvi commented 8 years ago

Currently splits longer sentences into several sequences.