You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Natural language generation from structured data #237
does ML.NET have a capability to extract text out of structured data ? Say for example , we have a stock data of a company over some years. I want to generate text based on a template for YoY Analysis.
Its text generation, I am talking about. Say I have sales for 2017 as 4M $ and 2018 its 5M$. I want to be able to generate text based on a template some think like this. " The sales increased from 4M $ to 5M $ by xyz percentage.".
closing as more info has not been provided since last year. Also as @zeahmed points this does not need ML. You seem to want to do some intelligent text generation using ML which can be achieved by using our transforms, trainers and some custom code that uses predictions results to generate text.
Activity
zeahmed commentedon May 24, 2018
Thanks for the question. Can you please elaborate more on this?
I don't understand if you want to do text extraction or generation?
as-ghub commentedon May 24, 2018
Its text generation, I am talking about. Say I have sales for 2017 as 4M $ and 2018 its 5M$. I want to be able to generate text based on a template some think like this. " The sales increased from 4M $ to 5M $ by xyz percentage.".
zeahmed commentedon May 29, 2018
As far as I understand from the example above, this problem can be solved without using machine learning at all.
Can you explain bit more why this problem would need machine learning?
fix for defaulting Averaged Perceptron # of iterations to 10 (dotnet#237
code migration from machinelearning-automl to machinelearning repo fe…
codemzs commentedon Jun 30, 2019
closing as more info has not been provided since last year. Also as @zeahmed points this does not need ML. You seem to want to do some intelligent text generation using ML which can be achieved by using our transforms, trainers and some custom code that uses predictions results to generate text.
fix for defaulting Averaged Perceptron # of iterations to 10 (dotnet#237
Added onnx export functionality for LpNormNormalizingTransformer (#4161)
Added Onnx export functionality to PCATransformer (#4188)