Just some templates. Easy to use and interpet. I programmed for 3 months in 2016 for my bachelor thesis, and then from february 2018 on.
So far created neural net templates:
- Cat/dog Image classifier (Convolutional Network) 92% val_acc.
- Google Stock and Bitcoin trend price predictor (Recurrent Neural Network)
- Bank Customer Churn Analysis(Artificial Neural Network w. backprop)
- Reinforcement Neural Net using apriori algorithm to detect preffered ads of person
- ...
Other templates include:
- LDA and PCA dimensionality reductions
- Logistic Regression
- Naive Bayes
- Decision Tree
- Random Forest
- Support Vector Machine
- Non Linear SVM
- Gaussian SVM
- K-Nearest Neighbors
- Natural Language Processing
- Hierarchical Clustering
- K-Means Clustering
- Parameter selection: GridSearchCV
- Model selection: K-crossfold validation
#Cat/dog Image classifier (Convolutional Network) 92% val_acc.