Skip to content

BKaiwalya/Deep-Learning_Human-Activity-Recognition

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits

ConfusionMatrix

ConfusionMatrix

 
 

HAR_ConfusionMatrix.py

HAR_ConfusionMatrix.py

 
 

HAR_DataPipeline.py

HAR_DataPipeline.py

 
 

HAR_HyperparameterOptimization.py

HAR_HyperparameterOptimization.py

 
 

HAR_model.py

HAR_model.py

 
 

HyperparameterTuning_Logs

HyperparameterTuning_Logs

 
 

README.md

README.md

 
 

TechicalPaper_Human Activity Recognition.pdf

TechicalPaper_Human Activity Recognition.pdf

 
 

TrainingLogs

TrainingLogs

 
 

Repository files navigation

Human Activity Recognition

Paper : Paper_Human Activity Recognition

Results:

  1. Data Preprocessing
  2. Input Pipeline
  3. Model architecture
  4. Metrics
    • Confusion Matrix
    • ROC AUC Curve
  5. Hyperparamter Optimization
    • Neurons in LSTM layer
    • Optimizer
    • Dropout rate

Results obtained so far:

Accuracy Test Data Set : 77.23 %

1. Training Data Visualization

Acc_TrainData

Gyro_TrainData

2. Test Data Visualization

Acc_TestData Gyro_TestData

3. Posture Data Visualization

WALKING

acceleration signals of experiment 1 when user 1 was performing activity_ 1(WALKING)

gyroscope signals of experiment 1 when user 1 was performing activity_ 1(WALKING)

WALKING_UPSTAIRS

gyroscope signals of experiment 1 when user 1 was performing activity_ 2(WALKING_UPSTAIRS)

acceleration signals of experiment 1 when user 1 was performing activity_ 2(WALKING_UPSTAIRS)

WALKING_DOWNSTAIRS

acceleration signals of experiment 1 when user 1 was performing activity_ 3(WALKING_DOWNSTAIRS)

gyroscope signals of experiment 1 when user 1 was performing activity_ 3(WALKING_DOWNSTAIRS)

SITTING

acceleration signals of experiment 1 when user 1 was performing activity_ 4(SITTING)

gyroscope signals of experiment 1 when user 1 was performing activity_ 4(SITTING)

STANDING

acceleration signals of experiment 1 when user 1 was performing activity_ 5(STANDING)

gyroscope signals of experiment 1 when user 1 was performing activity_ 5(STANDING)

LAYING

acceleration signals of experiment 1 when user 1 was performing activity_ 6(LAYING)

gyroscope signals of experiment 1 when user 1 was performing activity_ 6(LAYING)

STAND_TO_SIT

acceleration signals of experiment 1 when user 1 was performing activity_ 7(STAND_TO_SIT)

gyroscope signals of experiment 1 when user 1 was performing activity_ 7(STAND_TO_SIT)

SIT_TO_STAND

acceleration signals of experiment 1 when user 1 was performing activity_ 8(SIT_TO_STAND)

gyroscope signals of experiment 1 when user 1 was performing activity_ 8(SIT_TO_STAND)

SIT_TO_LIE

acceleration signals of experiment 1 when user 1 was performing activity_ 9(SIT_TO_LIE)

gyroscope signals of experiment 1 when user 1 was performing activity_ 9(SIT_TO_LIE)

LIE_TO_SIT

acceleration signals of experiment 1 when user 1 was performing activity_ 10(LIE_TO_SIT)

gyroscope signals of experiment 1 when user 1 was performing activity_ 10(LIE_TO_SIT)

STAND_TO_LIE

acceleration signals of experiment 1 when user 1 was performing activity_ 11(STAND_TO_LIE)

gyroscope signals of experiment 1 when user 1 was performing activity_ 11(STAND_TO_LIE)

LIE_TO_STAND

acceleration signals of experiment 1 when user 1 was performing activity_ 12(LIE_TO_STAND)

gyroscope signals of experiment 1 when user 1 was performing activity_ 12(LIE_TO_STAND)

4. Histogram: Training Examples per Activity

5. Metrics

Metrics_HAR

6. Confusion Matrix

Confusion matrix

7. ROC-AUC Curve

About

No description or website provided.

Topics

deep-learning time-series rnn lstm-neural-networks accelerometer-data gyroscope-data

Resources

Readme
Activity

Stars

0 stars

Watchers

1 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages