Adarsh SM

Strong in design and integration with intuitive problem-solving skills. Proficient in PYTHON, MACHINE LEARNING and SQL. Passionate about implementing and launching new projects. Ability to translate business requirements into technical solutions. Looking to start the career as an entry-level software engineer with a reputed firm driven by technology.

1.Human activity detection using deep learning:-

• I was able to build a model which can intelligently recognize the action sequence in video and can predict the labels. We have used the UCF-50 dataset and chose ten classes to train the model.
• I used two approaches which was LRCN(Long-term Recurrent Convolutional Network) architecture which combination of CNN and RNN.The second approach was Convlstm architecture.It was observed that LRCN had accuracy of 88% with loss of 0.49 and also took shorter time to train the model.
• skills used:-Deep Learning algorithms, Opencv, Keras, Tensorflow, Numpy,Matplotlib,Python.

2.Sales Forecasting Using Machine Learning :-

• my role in the project was backend and i was able to build model(xgboost regressor) which has capability to predict the future sales with accuracy more than 90%.
• skills used:-Xgboost regressor,minmax normalization(feature scaling),Python.