-
Notifications
You must be signed in to change notification settings - Fork 219
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Tf stage #71
Tf stage #71
Conversation
90c481c
to
8e62ba0
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good, only one minor thing about the models file.
model_ = tflite::FlatBufferModel::BuildFromFile(config_->model_file.c_str()); | ||
std::cout << "TfStage: Loaded model " << config_->model_file << std::endl; | ||
model_->error_reporter(); | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If there is an error (e.g. model file missing), do we need some kind of exception here?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, I'll fix that.
… stages The TfStage class provides a certain amount of TFLite boilerplate, and handles various tasks like fetching the streams, cropping the image to the right size, and so on. It also calls the interpreter asynchronously and provides an API for derived classes to implement to interpret and use the results. Signed-off-by: David Plowman <[email protected]>
…lass Mostly the effect is just to remove code which is now provided in the TfStage base class. Signed-off-by: David Plowman <[email protected]>
…lass Mostly this just involves the removal of code that is now in the base class. Signed-off-by: David Plowman <[email protected]>
…ion_tf.json This is the name it should have to match everything else. Signed-off-by: David Plowman <[email protected]>
8e62ba0
to
fe190e1
Compare
@naushir
This set of patches adds the new TfStage base class which makes the addition of new TFLite-based post-processing stages easier. The first of the commits adds the new TfStage class implementation.
Subsequently there's one patch that refactors the Object Classifier to use the new base class, and another that further refactors the Pose Estimator in the same way. (There's a final small file renaming patch that just fixes up an incorrect JSON file name.,)