- synthetic_data is the source code for Sec. II
- SensorFormer is the source code and data for our SensorFormer model
- one2one-w-val.py: one2one calibration scheme
- many2one-w-val.py: many2one calibration scheme
- many2many-w-val.py: many2many calibration scheme
- new_data/ : data used for the model
- result/ : results
- loss: loss functions used for the model
- attention_map: attention map data used for the experiment section
- SensorFormer_Lite is the source code and data for SensorFormer Lite model
- many2many_lite_random.py: input downsampling by random selection
- many2many_lite_mean.py: input downsampling by using the mean value
- many2many_lite_soft.py: input downsampling by weighted embeddings
- new_data/ : data used for the model
- result/ : results
- loss: loss functions used for the model
- attention_lite: attention map data used for the experiment section
- Arduino_model shows the code and steps to convert the trained model to code running on Arduino.
buptycz / sensorformer Goto Github PK
View Code? Open in Web Editor NEWThis project forked from shuowang-ai/sensorformer