Closed hua812586 closed 9 months ago
This has been rewritten and it will come soon, but for now here is dirty fix. You can use load_sample func from #9 and in plotter's draw_last_sample substitute:
last_sample = self.__get_previos_data(gesture)
with last_sample = load_sample(gesture)
.
plotter = Plotter(queue.Queue())
plotter.show()
plotter.draw_last_sample(<path>)
Or even easier, you can just substitue at line 85:
last_sample = Logger.get_last_sample(gesture)
if last_sample[-1] == '/':
return None
df = pd.read_csv(last_sample).reset_index().values
with
df = pd.read_csv(gesture).reset_index().values
and then again, instead of passing the gesture, you pass desired path to plotter.draw_last_sample
method.
Okay, I'll try both of those. Thank you very much for your clarification.
Hello @vilari-mickopf. I'm bothering you again. I modify the characteristics of the training (mainly in get_all_data-->get_data) and the monitored data (mainly in do_predict-->__predict_thread) on the basis of your code, but when plotting, obj is not displayed properly, what is this because? (I commented out the x, y handling in the plot_detected_objs and __get_previos_data functions).
Sorry for delay, repo is now updated and supports offline plotting. Have a look at main in mmwave_gesture/utils/plotter.py for an example. Data and models are now hosted on dropbox, so follow readme on how to get them (additional data is also recorded).
Hello @vilari-mickopf. I'm sorry to bother you again. I have a new problem, that is, how to display the data in the mmwave\data\ file as dynamically as in the plotter through offline mode.
Because I want to verify the correctness of the data I collected on the other radar boards in this way.