From 03011e1ff83c3425e186182067a87c7875b2e057 Mon Sep 17 00:00:00 2001 From: pierre-cau <pcaupro@gmail.com> Date: Fri, 15 Nov 2024 21:19:08 +0100 Subject: [PATCH] knn image --- README.md | 2 +- results/{knn.png => knn_accuracy.png} | Bin 2 files changed, 1 insertion(+), 1 deletion(-) rename results/{knn.png => knn_accuracy.png} (100%) diff --git a/README.md b/README.md index 5cde773..bd0e890 100644 --- a/README.md +++ b/README.md @@ -39,7 +39,7 @@ The code linked to the K-Nearest Neighbors algorithm is in the `src/utils/knn.py For a `split_factor` of 0.9, here is the accuracy of the K-Nearest Neighbors algorithm for different values of `k`: - + We notice that the best value of `k` is about 30%. **Thus, we can conclude that such a method is not very efficient for this application.** diff --git a/results/knn.png b/results/knn_accuracy.png similarity index 100% rename from results/knn.png rename to results/knn_accuracy.png -- GitLab