I’m encountering difficulties in the use of the “Epic epoch” script. I’ve seen questions similar to mine but I haven’t found a solution to my problem.
I would like to be able to work on a specific period of my corpus.
- For this, I used the “Period slicer” script as follows: [1940:2000];[2001:2005];[2006:2010];[2011:2015];[2016:2020]
- Then, I wanted to use the “Epic Epoch” script using “Custom period” with the number of periods = 5.
I get a different graph than the one obtained with “Epic Epoch” without pre-cutting my corpus. I conclude that the cutting has been taken into account. However I obtain a graph with 4 periods delimited by 10 vertical lines with 0; 0.5; 1; etc. I have the same result as with “Epic Epoch”. I don’t understand what this corresponds to.
Thank you in advance for your help!
I fact, what you are describing, is not so fare than what we are expecting with Epic Epoc.
In Period slicer script:
- when adding the custom periods defined as followed: [1940:2000];[2001:2005];[2006:2010];[2011:2015];[2016:2020]
- it will divide your corpus in five partitions (the custom periods).
Each period will received a number, starting from 0, up to 4 in your case. This is in fact how the Period slicer script is working.
So in the bump graph produced by Epic Epoc, click in the Dynamics option panel, and click on the on Custom period parameter (without changing nothing else in that option panel), and you should have what you want:
- a bump graph, with 5 periods, based on what you have defined with period slicer
- each integer are the periods : 0 for [1940:2000], 1 for [2001:2005] …
- do not look at the intermediary lines
If you are not satisfied by the labels, which is understandable, do not hesitate to use the green button (bottom left side of the Bump graph) to:
- remove the labels of the intermediary lines (0.5, 1.5 …)
- and replace the labels of the period by your couple of years
I hope it helps!
Thank you very much for your answer!
Indeed, I understand better the graph obtained with Epic Epoc. In my first reading, I took each separation as an indication of the limits of the periods (0 = 1940 on 1 = 2000, 2 = 2001, etc.), which did not make sense with the distance between each separation.
Everything is clearer, thank you very much for your help!