Maskrcnn-benchmark: Inference on new data set model

Created on 9 Nov 2018  ยท  1Comment  ยท  Source: facebookresearch/maskrcnn-benchmark

โ“ Questions and Help

What modifications should be made to the inference example in order to use a model trained on a new dataset? I trained a model using my own data and have multiple *.pth models created, but I don't see anywhere in the inference example config yaml to point to a .pth model. Thanks

question

Most helpful comment

Hi,

You can perform inference on your new model by changing the MODEL.WEIGHT argument in the config to point to your trained model. You can for example use the command-line overrides for that:

python tools/test_net.py --config-file path_to_your_cfg.yaml MODEL.WEIGHT path/to/your/model.pth

I'm closing this issue, but let us know if you have more questions;

>All comments

Hi,

You can perform inference on your new model by changing the MODEL.WEIGHT argument in the config to point to your trained model. You can for example use the command-line overrides for that:

python tools/test_net.py --config-file path_to_your_cfg.yaml MODEL.WEIGHT path/to/your/model.pth

I'm closing this issue, but let us know if you have more questions;

Was this page helpful?
0 / 5 - 0 ratings

Related issues

KuribohG picture KuribohG  ยท  3Comments

adityaarun1 picture adityaarun1  ยท  3Comments

CF2220160244 picture CF2220160244  ยท  3Comments

Jinksi picture Jinksi  ยท  3Comments

botcs picture botcs  ยท  3Comments