iDML
D's Machine Learning is a machine learning toolkit for python,focus on rightness but efficiency
all code is based on numpy and scipy
Code Files
./dml/NN -the code of Neural NetWorks
./dml/LR -Logistic Regression,actually It's softmax
./dml/DT -Decision Tree , CART algorithm
./dml/ClUSTER -some cluster algorithm,inculde kmeans \ kmedoids \ spectralCluster \ Hierarchical Cluster
./dml/ADAB -the adaboost algorithm
./dml/KNN -the k-Nearest Neighbor algorithm(kd-tree BBF implementing)
./dml/NB -the naive Bayesian support both continuous and descrete features
./dml/SVM -the basic binary Support Vector Machine
./dml/CNN -the simple Convolutional Neural Networks
./dml/CF -some Collaborative Filtering Algorithm implement,include item-based \ SVD \ RBM
./dml/tool -include some basic tools for computing
./test/ -include some test code for DML
Class Format
all class can be used in this way:(LR for example)
but there is still some different Initialization parameters in different class,also the predict function
sorry for this but most class use pred() and NN use nnpred(),I may formalize them in the future
a = LRC(train_images,trian_labels,nor=False)
a.train(200,True)
pred = a.predict(test_images)
for the input X and y ,X must be a N*M matrix and y is a vector length M
where N is the #feature and M is #training_case
for the cluster method,you can use a.labels or a.result() to get the final result
Install
DML is based on numpy,scipy,matplotlib .you should install them first
This packages uses setuptools, which is the default way of installing python modules. The install command is:(sudo is required in some system)
python setup.py build
python setup.py install
Warning
-
only python 2 is supported,sorry for the python 3 user.
-
some method from numpy and scipy will report warning because of their version