updated README.md on how to make changes

This pull request enhances the README documentation for the repository by providing clearer instructions for downloading GGML (Generative Grasping Machine Learning) files. This update aims to improve the user experience when working with GGML files, emphasizing efficient download practices and integration with Python for inference.

README.md

@@ -123,6 +123,22 @@ If you want to have a chat-style conversation, replace the `-p <PROMPT>` argumen
 
 Further instructions here: [text-generation-webui/docs/llama.cpp-models.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/llama.cpp-models.md).
 
+## How to make changes. 
+
+Downloading GGML files from repositories can be done in a more efficient way than using git clone. Here's how:
+
+1. Avoid Git Clone: It's not recommended to use git clone for GGML repositories as it downloads all files, resulting in unnecessary data storage.
+2. Use wget or Download Client: Instead, use the wget command or any download client to directly grab the specific file you need. This prevents excessive data usage and storage.
+3. Python Code Option: If you prefer Python, you can use the hf_hub_download function from huggingface_hub. Here's an example:
+
+```
+from huggingface_hub import hf_hub_download
+hf_hub_download(repo_id='TheBloke/Llama-2-7B-Chat-GGML', filename='llama-2-7b-chat.ggmlv3.q4_K_M.bin')
+#This code snippet downloads the specified GGML file directly.
+```
+
+4. Inference from Python: If you plan to perform inference from Python, the ctransformers Python library is a convenient option. It handles automatic file downloads for GGML inference.
+
 <!-- footer start -->
 ## Discord