search

nodefluxio / vortex

A Deep Learning Model Development Framework for Computer Vision
27 stars 6 forks source link

[BUG] output format is not defined #66

Closed internnos closed 4 years ago

internnos commented 4 years ago

Describe the bug NameError: name 'output_format' is not defined even though I had given the extra parameter self.task and self.output_format inside the model at nodeflux/face_detection/retinaface.py

To Reproduce Steps to reproduce the behavior: clone git@gitlab.com:nodefluxio/rexxar-rg/inference-repository/face-detection.git into branch 2-export-to-vortex run export.py

Expected behavior

Screenshots image

Additional context Done on v0.2

alphinside commented 4 years ago

where can i get the weights file? @amajidsinar

triwahyuu commented 4 years ago

can you send me the model definition and your export.py code? @amajidsinar

internnos commented 4 years ago

@triwahyuu @alphinside sorry I forgot to mention that you need to run sh retrieve_weight to get the model. I just tried it and it worked without problem. But if it's not working you can download it at s3://nodeflux-data/nodeflux-zoo/object-detection/wide-area-face-detection_-_1-class/v1/mobilenet0.25_Final.pth

@triwahyuu you can clone from here git@gitlab.com:nodefluxio/rexxar-rg/inference-repository/face-detection.git into branch 2-export-to-vortex as I've mentioned above

Anyway, many thanks for the quick reply

triwahyuu commented 4 years ago

so is it solved?

internnos commented 4 years ago

@triwahyuu not yet

triwahyuu commented 4 years ago

okay, on it

internnos commented 4 years ago

@triwahyuu cool, thanks a bunch