Icon

TARELA-LSTM

<p>TARELA-LSTM</p>

TARELA-LSTM

sort:1. by Country (asc)2. Date_reported (asc)
Sorter
load:New_cases (time series multi country)
Excel Reader
for each country
Group Loop Start
preprocessing(STAGE 1)
Python Script
list of countries
GroupBy
partitioning:train (inner),validation (in-sample),test (out-of-sample)include:carry over 7 days
Python Script
config :objective function
Table Creator
RevIN-LSTM baseline prediction output (output-0)
Excel Reader
config :objective function(FULL - conservative / C0)
Table Creator
TEST set(out-of-sample set)
Row Filter
Table Row to Variable
STAGE 2
Recursive Loop Start
prepare output
Metanode
define experiment GENERAL attribute:country, model, other atributes
Table Creator
for each window(STAGE 1)
Recursive Loop Start
config :objective function(config C1)
Table Creator
config :objective function(MSE-only)
Table Creator
config :objective function(config C2)
Table Creator
TRAIN New_cases(based on window)
Row Filter
Transition Label Builder
Python Script
config :objective function(config C3)
Table Creator
TEST New_cases(based on window)
Row Filter
model TARELA-LSTM
Python Script
save output
reset RowID
RowID
Table Row to Variable
loop: window(end of STAGE 1)
Recursive Loop End
Window Table Generator(#windows =10)
Python Script
based on:currentIteration
Current Window
sequence builder
Python Script
TRAIN set
Row Filter
loop: window
Recursive Loop End
path:relative to current workflow
prepare filename
loop: country
Loop End
filter:Country
Row Filter

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: adria
Created at: 2026-04-22
On NodePit since: 2026-08-13
Last update: 2026-08-13
Created with KNIME version: v5.8.3

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!