Closed myunggiM closed 5 years ago
isn't it using wrong factor? in iterator_utils.py? in line 168~173
src_tgt_dataset = src_tgt_dataset.map( lambda src, tgt: (src, tf.concat(([tgt_sos_id], tgt), 0), tf.concat((tgt, [tgt_eos_id]), 0)), num_parallel_calls=num_parallel_calls).prefetch(output_buffer_size)
input lambda factor is 2 but number of output is 3 therefore i think it( src,tgt ) is edited like src, tgt_in , tgt_out
so it'll be src_tgt_dataset = src_tgt_dataset.map( lambda src, tgt_in , tgt_out: (src, tf.concat(([tgt_sos_id], tgt), 0), tf.concat((tgt, [tgt_eos_id]), 0)), num_parallel_calls=num_parallel_calls).prefetch(output_buffer_size)
isn't it? plz give me feedback..
isn't it using wrong factor? in iterator_utils.py? in line 168~173
Create a tgt_input prefixed with and a tgt_output suffixed with .
src_tgt_dataset = src_tgt_dataset.map( lambda src, tgt: (src, tf.concat(([tgt_sos_id], tgt), 0), tf.concat((tgt, [tgt_eos_id]), 0)), num_parallel_calls=num_parallel_calls).prefetch(output_buffer_size)
input lambda factor is 2 but number of output is 3
therefore i think it( src,tgt ) is edited like src, tgt_in , tgt_out
so it'll be src_tgt_dataset = src_tgt_dataset.map( lambda src, tgt_in , tgt_out: (src, tf.concat(([tgt_sos_id], tgt), 0), tf.concat((tgt, [tgt_eos_id]), 0)), num_parallel_calls=num_parallel_calls).prefetch(output_buffer_size)
isn't it? plz give me feedback..