diff --git a/README.md b/README.md
index bbd46debe3f7fc815b6e668074ad3b5ca8681434..2f3f298329e027a498a2b5911098162a9a122295 100644
--- a/README.md
+++ b/README.md
@@ -9,7 +9,7 @@ The two parts are trained together in a process called adversarial training. The
 
 GANs have many applications, including image synthesis, text generation, and video generation. They have been used to create photorealistic images, to generate music, and even to create new molecules with desired properties.
 
-## Training 1
+### Training 1
 
 
 In this section, a **Generative Adversarial Network (GAN)** was utilized to produce novel handwritten digits by utilizing images from the **MNIST** database as input. The **DCGAN tutorial of Pytorch** was employed to achieve this goal, which generates new images of celebrities by analyzing numerous real celebrity images. To adjust the tutorial to the MNIST database, the primary modification in its implementation involved changing the number of channels in the training images from 3 (RGB) to 1 (black and white).