T-values Vs. P-values

Question

I have a classification Python program that compares between four algorithms in both P-values and T-values. A comparison was conducted between the predicted labels and the original labels. For each of the algorithms, I ran them three times, capturing their values, p-values, and T-tests. At the end, I used boxplots to plot the values where i got something like these:

**My question is:

1. Which one of the algorithms do you consider the best (in t-values and p-values) and why?**

I used the following code:

plt.figure(figsize = (7, 7))
plt.boxplot([totalT['poly'], totalT['rbf'], totalT['linear'], totalT['gf']])
plt.xticks(np.arange(1, 5), kernels)
plt.title('T values for each svm kernel')
plt.xlabel('SVM kernel')
plt.ylabel('T values Rate')
plt.ioff()
plt.savefig('images/tValues.png')
plt.show()