search

CUNY-CL / yoyodyne

Small-vocabulary sequence-to-sequence generation with optional feature conditioning
Apache License 2.0
30 stars 17 forks source link

Issue in HardAttentionLSTM #258

Closed alejandromunoznavarro closed 1 week ago

alejandromunoznavarro commented 2 weeks ago

There is an issue in hard_attention.py. get_decoder is missing the parameter embeddings when it calls modules.lstm.ContextHardAttentionLSTMDecoder (lines 413 and 433). This could be fixed adding the line embeddings=self.embeddings in both calls.

kylebgorman commented 1 week ago

Thanks for the report, should be an easy fix. I'll take a stab at it.

kylebgorman commented 1 week ago

I believe I have taken care of this but let me know if anything lingers.