def remap()

in tools/taobao_prepare.py [0:0]

• 19 lines of code
• 2 McCabe index (conditional complexity)

def remap(df):
    item_key = sorted(df['iid'].unique().tolist())
    item_len = len(item_key)
    item_map = dict(zip(item_key, range(item_len)))

    df['iid'] = df['iid'].map(lambda x: item_map[x])

    user_key = sorted(df['uid'].unique().tolist())
    user_len = len(user_key)
    user_map = dict(zip(user_key, range(item_len, item_len + user_len)))
    df['uid'] = df['uid'].map(lambda x: user_map[x])

    cate_key = sorted(df['cid'].unique().tolist())
    cate_len = len(cate_key)
    cate_map = dict(zip(cate_key, range(user_len + item_len, user_len + item_len + cate_len)))
    df['cid'] = df['cid'].map(lambda x: cate_map[x])

    btag_key = sorted(df['btag'].unique().tolist())
    btag_len = len(btag_key)
    btag_map = dict(zip(btag_key, range(user_len + item_len + cate_len, user_len + item_len + cate_len + btag_len)))
    df['btag'] = df['btag'].map(lambda x: btag_map[x])

    print(item_len, user_len, cate_len, btag_len)
    return df, item_len, user_len + item_len + cate_len + btag_len + 1 #+1 is for unknown target btag