def get_data():
df = dataio.read_process("datasets/ml-20m/ratings.csv", sep=",")
rows = len(df)
df = df.iloc[np.random.permutation(rows)].reset_index(drop=True)
split_index = int(rows * 0.9)
df_train = df[0:split_index]
df_test = df[split_index:].reset_index(drop=True)
return df_train, df_test
def get_data():
df = dataio.read_process("./tmp/movielens/ml-1m/ratings.dat", sep="::")
rows = len(df)
df = df.iloc[np.random.permutation(rows)].reset_index(drop=True)
split_index = int(rows * 0.9)
df_train = df[0:split_index]
df_test = df[split_index:].reset_index(drop=True)
return df_train, df_test, rows
def get_data():
df = dataio.read_process(r"D:\Users\fuzzhang\software\tensorflow\TF_Recommend_Basic\TF_Recommend_Basic\TF-recomm\ratings.dat", sep="::")
rows = len(df)
df = df.iloc[np.random.permutation(rows)].reset_index(drop=True)
split_index = int(rows * 0.9)
df_train = df[0:split_index]
df_test = df[split_index:].reset_index(drop=True)
return df_train, df_test
def get_data():
df = dataio.read_process("/tmp/movielens/ml-1m/ratings.dat", sep="::")
rows = len(df)
df = df.iloc[np.random.permutation(rows)].reset_index(drop=True)
split_index = int(rows * 0.9)
df_train = df[0:split_index]
df_test = df[split_index:].reset_index(drop=True)
return df_train, df_test
def get_data():
df = dataio.read_process(
"/Users/chengyao/Projects/netease_spider/music_data/records_movielens_like.csv", sep=",")
rows = len(df)
df = df.iloc[np.random.permutation(rows)].reset_index(drop=True)
split_index = int(rows * 0.9)
df_train = df[0:split_index]
df_test = df[split_index:].reset_index(drop=True)
return df_train, df_test
def get_data():
df = dataio.read_process("/tmp/movielens/ml-1m/ratings.dat", sep="::")
wt = pd.read_csv("totalfortest2.csv")
rows = len(df)
#print(wt.shape)
df = df.iloc[np.random.permutation(rows)].reset_index(drop=True)
split_index = int(rows * 0.9)
df_train = df[0:split_index]
df_test = df[split_index:].reset_index(drop=True)
return df_train, df_test, wt
def get_data():
df = dataio.read_process("/tmp/movielens/ml-1m/ratings.dat", sep="::")
rows = len(df)
df = df.iloc[np.random.permutation(rows)].reset_index(drop=True)
split_index = int(rows * 0.9)
df_train = df[0:split_index]
df_test = df[split_index:].reset_index(drop=True)
maxtime = max(df['st'])
ut_mean = df.groupby(['user'])['st'].mean()
return df_train, df_test, maxtime, ut_mean
def get_data():
df = dataio.read_process("data/user_basket_size.csv", sep=",")
df['group_indicator'] = (df.ix[:, 0] != df.ix[:, 0].shift(-1)).astype(int)
df_train = df.loc[df.group_indicator == 0]
df_train = df_train.drop('group_indicator', axis=1)
df_test = df.loc[df.group_indicator == 1]
df_test = df_test.drop('group_indicator', axis=1)
df = df.drop('group_indicator', axis=1)
return df_train, df_test
def get_data():
df = dataio.read_process("/Users/xinghailong/Documents/workspace/my/DMInAction/src/tesnsorflow/recommend/ml-1m/test.dat", sep="::")
rows = len(df)
# 打乱顺序
# np.permutation(rows) 随机生成rows内的一维数组：
# https://docs.scipy.org/doc/numpy/reference/generated/numpy.random.permutation.html
# 并且重置索引
df = df.iloc[np.random.permutation(rows)].reset_index(drop=True)
# 切分数据集，90%用于训练集，10%用于测试集
split_index = int(rows * 0.9)
df_train = df[0:split_index]
df_test = df[split_index:].reset_index(drop=True)
return df_train, df_test
def get_data():
'Grab data, compute implicit matrix, and do train-test split'
# Grab data using dataio functions
df = dataio.read_process(
"../data_cleaning/data_for_CF/user_item_rating_fac.csv", sep=",")
# Compute implicit matrix
implicit_mat = df.pivot(index='user', columns='item',
values='rate').notnull().as_matrix().astype(float)
# Perform data shuffle
rows = len(df)
sample_index = np.random.permutation(rows)
#df = df.iloc[sample_index].reset_index(drop=True)
# Train-test split
split_index = int(rows * 0.9)
#df_train = df[0:split_index]
df_train = df
df_test = df[split_index:].reset_index(drop=True)
return df_train, df_test, implicit_mat
