I want to export the plot points of a cox regression survival plot so i can use it in Graphpad Prism to make a nice graph out of it.
Question
How can i export the plot points?
is there maybe an easier way to export such a graph to edit it in graphpad?
You should add a save subcommand to your analysis syntax:
/SAVE=SURVIVAL
After running the analysis you will find new column(s) containing the survival data which you can then export and plot elsewhere.
If you look in the Menu for the analysis you can see a button called - click it to see all options of new variables to save, maybe you can find other data useful.
Related
I would like to create a forest plot for the meta-regression outputs using metafor. There are several effect modifiers in the regression model I created. I am not sure if I can visualize them in one forest plot. I would like the forest plot to look as where ylab would show all the effect modifiers and xlab would show log response ratio. Would that be possible and if it is, could you please give an example?
I'm trying to plot my model in Keras, like this:
# Plot model graph
tf.keras.utils.plot_model(model, to_file='Model1.png')
from IPython.display import Image
Image(retina=True, filename='Model1.png')
Which I get the following result: my model
But, I've seen somewhere in the internet, that someone plotted his model, like this: model I need
How can I change my code to plot like that? With the input/output information of each layer of my model?
You can use the parameter show_shapes=True.
from the tf.keras documentation:
show_shapes: whether to display shape information.
(have a closer look here: https://www.tensorflow.org/api_docs/python/tf/keras/utils/plot_model )
Im looking for the name of the library capable of making this plot.
The forecasting bit is I'm what I'm interested. Is it Forecast library?
Thanks.
Yes, the time-series plot is from the forecast library. You get that sort of plot when you plot a forecast object. Look at ?forecast. For example:
install.packages('forecast')
library(forecast)
fit <- StructTS(WWWusage,"level")
plot(forecast(fit))
The rest of the fancy stuff i.e. interactivity is most likely shiny magic, but I think you are interested only in the main plot.
Does anyone know if it is possible to view the data used for plotting the surface regression in the scatter3d package in R? What I want is to be able to somehow view the formula (or function) for the surface regression that is be applied to my data points in the scatter3d package in R.
Edit: I'm guessing you're using scatter3d from the IPSUR package.
Per the help page,
"Value
scatter3d not return a useful value; it is used for its side-effect of creating a 3D scatterplot. indentify3d returns the labels of the identified points."
You might want to experiment with scatterplot3D over at CRAN to see if its returned values (functions) are of use.
You might be able to dig into the source of scatter3d and fork a version which returns the data.
I would like to produce some nice PCA plots in R. As usual, in R, there are several ways to perform a principal component analysis. I found so far 3 different ways of how to calculate your components and 3 ways of plotting them. I was wondering whether people who are familiar with these functions can give me some advise on the best combination of functions to produce the following plots:
Scores Plot
Loadings Plot
Histogram / Bar chart of the variances explained by each principal component
My research on functions and plots used for PCA in R resulted in:
Functions:
pca.xzy()
prcomp()
princomp()
dudi.pca()
Plot:
plot.pca (this one seems to belong to the function pca.xzy())
ggplot2
plot
biplot
I also found the following webpage:
http://pbil.univ-lyon1.fr/ade4/ade4-html/dudi.pca.html
And I was wondering if you can draw those circles and lines starting from each of the circle centers with one of the other functions mentioned above as the function dudi.pca from the ade4 package seems to be the most complicated one.
One question per question, please! There's psych package by William Revelle, see this and this. There's also a good tutorial here. Anyway...
for scores/loadings plot see pairs
histogram: see hist
So once again, what's your question actually? =)