From df10550a9fbedc09d7eddc30c0065dfe1764e5ca Mon Sep 17 00:00:00 2001 From: GuoKaku Date: Thu, 28 Sep 2023 04:02:25 +0000 Subject: [PATCH] update tutorial about timeF encoding --- tutorial/TimesNet_tutorial.ipynb | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tutorial/TimesNet_tutorial.ipynb b/tutorial/TimesNet_tutorial.ipynb index caaf0181..6534415e 100644 --- a/tutorial/TimesNet_tutorial.ipynb +++ b/tutorial/TimesNet_tutorial.ipynb @@ -194,7 +194,7 @@ "\n", " # find period by amplitudes: here we assume that the peroidic features are basically constant\n", " # in different batch and channel, so we mean out these two dimensions, getting a list frequency_list with shape[T] \n", - " # each element at pos t of frequency_list demeans the overall amplitude at frequency (t)\n", + " # each element at pos t of frequency_list denotes the overall amplitude at frequency (t)\n", " frequency_list = abs(xf).mean(0).mean(-1) \n", " frequency_list[0] = 0\n", "\n", @@ -1315,7 +1315,7 @@ " options:[s:secondly, t:minutely, h:hourly, d:daily, b:business days, w:weekly, m:monthly], you can also use more detailed freq like 15min or 3h')\n", " So you should check the timestep of your data and set 'freq' arg. \n", " After the time_features encoding, each date info format will be encoded into \n", - " a list, with each element demeaning the relative position of this time point\n", + " a list, with each element denoting the relative position of this time point\n", " (e.g. Day of Week, Day of Month, Hour of Day) and each normalized within scope[-0.5, 0.5]\n", " '''\n", " data_stamp = time_features(pd.to_datetime(df_stamp['date'].values), freq=self.freq)\n",