name 'WINDOW_SIZE' is not defined

[fattypiggy]

inputs = Input(shape=(WINDOW_SIZE, NUMBER_FEATURES, 1,))

can you tell me how to fix this?

[AntoBr96]

Hi, I think you forgot to declare the constant WINDOW_SIZE. Best, Antonio Il ven 13 nov 2020, 04:10 经纬 <notifications@github.com> ha scritto:

…

inputs = Input(shape=(WINDOW_SIZE, NUMBER_FEATURES, 1,)) can you tell me how to fix this? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/APQ2FOU3LVIJNU3TKEK63ATSPSPSJANCNFSM4TUCC35A> .

[fattypiggy]

Hi, I’m a beginner at this, so can you tell me The exactly numbers or what are these two numbers related to? Thanks very much. Best, William AntoBr96 <notifications@github.com>于2020年11月13日 周五下午3:23写道：

Hi, I think you forgot to declare the constant WINDOW_SIZE. Best, Antonio Il ven 13 nov 2020, 04:10 经纬 ***@***.***> ha scritto: > inputs = Input(shape=(WINDOW_SIZE, NUMBER_FEATURES, 1,)) > > can you tell me how to fix this? > > — > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub > <#1>, or > unsubscribe > < https://github.com/notifications/unsubscribe-auth/APQ2FOU3LVIJNU3TKEK63ATSPSPSJANCNFSM4TUCC35A > > . > — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEGUTKIKXNTOL3LECHOWN73SPTNGPANCNFSM4TUCC35A> .

-- Best regards, -- William Jing

[AntoBr96]

Hi, For sure I can. The WINDOW_SIZE refers to the history you want to pass to your architecture. If it is 100, for example, you will pass to the architecture all the ticks between t-99 and t. The NUMBER_FEATURES value refers to the number of limit order book’s levels. For a complete explanation please read: https://arxiv.org/pdf/2007.07319.pdf <https://arxiv.org/pdf/2007.07319.pdf> . For any academic work please always cite the same paper. Best, Antonio Briola

…

Il giorno 13 nov 2020, alle ore 08:29, 经纬 ***@***.***> ha scritto: Hi, I’m a beginner at this, so can you tell me The exactly numbers or what are these two numbers related to? Thanks very much. Best, William AntoBr96 ***@***.***>于2020年11月13日 周五下午3:23写道： > Hi, > > I think you forgot to declare the constant WINDOW_SIZE. > > Best, > Antonio > > Il ven 13 nov 2020, 04:10 经纬 ***@***.***> ha scritto: > > > inputs = Input(shape=(WINDOW_SIZE, NUMBER_FEATURES, 1,)) > > > > can you tell me how to fix this? > > > > — > > You are receiving this because you are subscribed to this thread. > > Reply to this email directly, view it on GitHub > > <#1>, or > > unsubscribe > > < > https://github.com/notifications/unsubscribe-auth/APQ2FOU3LVIJNU3TKEK63ATSPSPSJANCNFSM4TUCC35A > > > > . > > > > — > You are receiving this because you authored the thread. > Reply to this email directly, view it on GitHub > <#1 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AEGUTKIKXNTOL3LECHOWN73SPTNGPANCNFSM4TUCC35A> > . > -- Best regards, -- William Jing — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/APQ2FORH3WIJNFAS34CB4KDSPTN3ZANCNFSM4TUCC35A>.

[fattypiggy]

thanks for your advice, I have set
WINDOW_SIZE = 100
NUMBER_FEATURES = 1
and now It's runnable, thanks very much.
And there is another question: how can I use this returned model? I tried just run cnn_lstm_model() but it doesn't work.

[AntoBr96]

Hi, Once you compiled the model, you have to fit it and then you can evaluate the model on your test data. If you didn’t work with Deep Learning before you can read the following introductory tutorial: https://machinelearningmastery.com/tutorial-first-neural-network-python-keras/ <https://machinelearningmastery.com/tutorial-first-neural-network-python-keras/>

…

Il giorno 17 nov 2020, alle ore 03:29, 经纬 ***@***.***> ha scritto: thanks for your advice, I have set WINDOW_SIZE = 100 NUMBER_FEATURES = 1 and now It's runnable, thanks very much. And there is another question: how can I use this returned model? I tried just run cnn_lstm_model() but it doesn't work. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/APQ2FOT4HFR5E3TX6RWWE2LSQHNZRANCNFSM4TUCC35A>.

[fattypiggy]

thank you very much, you are so kind to tell me this patiently.

…

On Tue, Nov 17, 2020 at 3:28 PM AntoBr96 ***@***.***> wrote: Hi, Once you compiled the model, you have to fit it and then you can evaluate the model on your test data. If you didn’t work with Deep Learning before you can read the following introductory tutorial: https://machinelearningmastery.com/tutorial-first-neural-network-python-keras/ < https://machinelearningmastery.com/tutorial-first-neural-network-python-keras/> > Il giorno 17 nov 2020, alle ore 03:29, 经纬 ***@***.***> ha scritto: > > > thanks for your advice, I have set > WINDOW_SIZE = 100 > NUMBER_FEATURES = 1 > and now It's runnable, thanks very much. > And there is another question: how can I use this returned model? I tried just run cnn_lstm_model() but it doesn't work. > > — > You are receiving this because you commented. > Reply to this email directly, view it on GitHub < #1 (comment)>, or unsubscribe < https://github.com/notifications/unsubscribe-auth/APQ2FOT4HFR5E3TX6RWWE2LSQHNZRANCNFSM4TUCC35A >. > — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEGUTKPWN2PV4Y6RIGSACKLSQIQ25ANCNFSM4TUCC35A> .

-- Best regards, -- William Jing