'How to process multiple tensorflow models android
I am working on the app which will take the camera input and process with tensorflow model(for gesture detection) which will give the bounding box output. I have to take the output from the first model and should process the same to the second model(identify gesture). I could not find anything. please guide me in the right direction. Thanks in advance.
Solution 1:[1]
simple explanation is a prediction with target conditions preferred path. Q-networks may return more than action but expect values or pausing action that is because it can share loads.
When action is only X capable how do you tell the SpaceX to fires those enemies
How to make contast fastest way? It start firing since first rounds.
[ Sample ]:
predictions_1 = mainQ_outputs.predict(output_picture[0])
max_prediction_1 = np.argmax(predictions_1)
predictions_2 = mainQ_outputs.predict(output_picture[0])
max_prediction_2 = np.argmax(predictions_2)
predictions_3 = mainQ_outputs.predict(output_picture[0])
max_prediction_3 = np.argmax(predictions_3)
predictions_4 = mainQ_outputs.predict(output_picture[0])
max_prediction_4 = np.argmax(predictions_4)
### Max value may consider most active
max_predictions = np.max[ np.max(max_prediction_1), np.max(max_prediction_2),
np.max(max_prediction_3), np.max(max_prediction_4) ]
...
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|---|
| Solution 1 | Martijn Pieters |