Replies: 1 comment
-
Ok Sent from my BlackBerry 10 smartphone. From: Selorm KomlaSent: Monday, 21 August 2023 21:01To: mrdbourke/tensorflow-deep-learningReply To: mrdbourke/tensorflow-deep-learningCc: SubscribedSubject: [mrdbourke/tensorflow-deep-learning] notebook 10, model 3: Preparing the data. (Discussion #576)
I need some more clarification on the WINDOW_SIZE and HORIZON's operation in the data preparation. From my understanding, a window_size of 7 and Horizon of 1, produces a vector of length 7 for the window and scalar (vector of length of 1) for the Horizon. So when we used a window size of 30 and Horizon of 7 model model 3, I was expecting to see vectors of length 30 and 7 respectively for train and training and labels. However, I see it results in the same kind of length as used in model 2. That is, After using window_size of 30, the vector for training and test windows has a length of 30, but the labels still have a length of 1 instead of 7.
Why is this so?
—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>
[
{
***@***.***": "http://schema.org",
***@***.***": "EmailMessage",
"potentialAction": {
***@***.***": "ViewAction",
"target": "#576",
"url": "#576",
"name": "View Discussion"
},
"description": "View this Discussion on GitHub",
"publisher": {
***@***.***": "Organization",
"name": "GitHub",
"url": "https://github.com"
}
}
]
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I need some more clarification on the WINDOW_SIZE and HORIZON's operation in the data preparation. From my understanding, a window_size of 7 and Horizon of 1, produces a vector of length 7 for the window and scalar (vector of length of 1) for the Horizon. So when we used a window size of 30 and Horizon of 7 model model 3, I was expecting to see vectors of length 30 and 7 respectively for train and training and labels. However, I see it results in the same kind of length as used in model 2. That is, After using window_size of 30, the vector for training and test windows has a length of 30, but the labels still have a length of 1 instead of 7.
Why is this so?
Beta Was this translation helpful? Give feedback.
All reactions