Files
stree_datasets/notes.txt
2020-11-20 11:23:40 +01:00

44 lines
1.5 KiB
Plaintext
Executable File

iris.csv
vehicle.csv
wine.csv
glass.csv
heart.csv # spaces
breast # 683 vs 690 samples in dataset
diabetes # from Kaggle
fourclass # taken from libsvm samples TKH96a Tin Kam Ho and Eugene M. Kleinberg.
#Building projectable classifiers of arbitrary complexity.
#In Proceedings of the 13th International Conference on Pattern Recognition, pages 880-885, Vienna, Austria, August 1996.
segment # sparse libsvm para cargar X, y = np.load("data/segment.npy", allow_pickle=True) para leer .scale sklearn.datasets.load_svmlight_file
letter
sat (has to be satimage)
usps # kaggle import h5py
with h5py.File(path, 'r') as hf:
train = hf.get('train')
X_tr = train.get('data')[:]
y_tr = train.get('target')[:]
test = hf.get('test')
X_te = test.get('data')[:]
y_te = test.get('target')[:]
pendigits
protein # https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass.html#connect-4
dna # openml
connect 4 # https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass.html#connect-4
ijcnn1 # libsvm https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary.html
Datasets que hay que cargar X e y y son sparse matrices
=======================================================
(X, y= np.load("data/connect4.npy", allow_pickle=True))
connect4
fourclass
ijcnn1
protein
segment
vehicle
from svmlight_loader import (load_svmlight_file, load_svmlight_files,
dump_svmlight_file)
para enlazar matrices sparse:
from scipy.sparse import vstack
X = vstack((Xs, Xt))