I have trained a CNN model and I have applied 10 Fold Cross Validation because I don't have much data to train the classifier. Now I am unsure about how to visulize fold wise results. Please suggest some visualization charts or techniques to display fold wise results.
Do you only have one single model? If you were only mixing the data up (and not trying different parameter values) then any plot might not be very really useful. If performance is on the y-axis, what would be on the x-axis? The fold number itself doesn't really give any insight regarding the results (assuming random batch selection).
Are there some statistics you could compute for each of the 10 folds? If there is any metric you can compute for the x-axis, then using a simple line chart would help see how performance varies with that metric. For example, if you have images, you could compute the average pixel value, or the number of true positives/negatives in that batch (if that idea applies to your problem?). An example from Scikit Learn is this, where the gamma parameter of an SVM is altered on the x-axis:
Additionally, if you have several models (e.g. CNNs with varying numbers of layers / depth), you could use a nice box-whisker plot. Seaborn offers some nice variations. Here is one example that would show four different models:
You could make a box plot for a single model too, which wouldn't be so appealing, but might look like this:
Again, I'm not really sure what you would be putting on the axes.