To see the training curve using ner.train-curve

[Quanti (W)]

Hi,

By using the ner.train-curve I am able to see the result run, but how do I see the curve based on the documentation presents?https://prodi.gy/docs/workflow-named-entity-recognition

[ines (Ines Montani)]

The command gives you the accuracy numbers for each training experiment – for the docs, I just took the values and mapped them out as a graph. So you could probably do the same in a Jupyter notebook or Python script using something like matplotlib using the accuracy for the Y-axis and the number of examples for the X-axis.