Channels error while translating

[HongChow]

RuntimeError: Given groups=1, weight of size 64 1 3 3, expected input[30, 3, 60, 360] to have 1 channels, but got 3 channels instead

command line is : onmt_translate -data_type img \

           -model demo-model_step_100000.pt \
           -src_dir data/im2text/images \
           -src data/im2text/src-test.txt \
           -output pred.txt \
           -max_length 150 \
           -beam_size 5 \
           -gpu 0 \
           -verbose

[guillaumekln (Guillaume Klein)]

You had the same issue in:

Question about the input channel error

Hi, all Great Thanks to OpenNMT. However, I am a beginner of NLP related. When i try to translate image to latext (im2text) using a image screenshot from some paper , I got the error information. In detail, with "-image_channel_size 1 " at the end of translate.py command line , it is “Given groups=1, weight of size 64 3 3 3, expected input[1, 1, 97, 500] to have 3 channels, but got 1 channels instead” and “Given groups=1, weight of size 64 3 3 3, expected input[1, 4, 97, 500] to have 3 chan…

and

what is the requirement for the input image size?

Looks like the solution is similar?