Skip to content

Using the api myo-python to train a recurrent neural network to predict hand gestures

Notifications You must be signed in to change notification settings

brunorchaves/myo_predictor

 
 

Repository files navigation

Work in progress

Myo Predictor

Using the api myo-python (add link later) to train a recurrent neural network to predict hand gestures

The code paramDataset.py at myo_predictor/gesture_predictor/ is used to capture signals from the Myo armband and generate .csv files to make datasets for further analysis.

The code trainGesture.py at myo_predictor/gesture_predictor/ is a neural network implemenation designed to be trained with the .csv files generated with paramDataset.py and to create a model.

The code runModel.py, as the name says, is used to run the classification model in real time with the armband.

How to merge .csv into only one (windows):

  1. Browse to the folder with the CSV files.
  2. Hold down Shift, then right-click the folder and choose Copy as path.
  3. Open the Windows Command prompt.
  4. Type cd, press Space, right-click and select Paste, then press Enter.
  5. Type copy *.csv combined-csv-files.csv and Press Enter.

About

Using the api myo-python to train a recurrent neural network to predict hand gestures

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Jupyter Notebook 93.1%
  • Python 4.9%
  • C 1.2%
  • PureBasic 0.8%