Update README.md

README.md

@@ -1,24 +1,18 @@
+---
+license: apache-2.0
+title: 'BLLAMA: ALPACA with BLIP 2'
+sdk: gradio
+emoji: 🔥
+colorFrom: red
+colorTo: purple
+---
 ## 🦙🌲🤏 BLLAMA: A BLIP2 + ALPACA-LORA Pipeline
 
-# Setup
-1. Git clone this repository
-2. ```pip install -r requirements.txt```
-
 # Training
  This is just a pipeline involving the use of both ALPACA and BLIP-2, without any prior finetuning. You can refer to the details in ALPACA_LORA's repo [here](https://github.com/tloen/alpaca-lora) and the BLIP-2 training details on their GitHub page [here](https://github.com/salesforce/LAVIS/tree/main/projects/blip2). For the pipeline, I have used the BLIP-2 model found on HuggingSpace [here](https://huggingface.co/spaces/Salesforce/BLIP2)
 
-# Inference
-1. cd to the cloned repo
-2. Run ```python3 generate.py```
-
-# Sample of inference
-![My Image](Results.png)
-
-
-#TODO:
-1. Try to reduce VRAM Usage: It hits around 14GB of VRAM on the 7B Weights when combined with BLIP2
-2. Add ability for users to customise their prompts to BLIP-2 in Gradio. This can help finetune the context given from BLIP2 to ALPACA, improving accuracy of generated outputs
 
 
 ## Acknowledgements
-Once again, I would like to credit the Salesforce team for creating BLIP2, as well as tloen, the original creator of alpaca-lora.
+Once again, I would like to credit the Salesforce team for creating BLIP2, as well as tloen, the original creator of alpaca-lora. I would also like to credit Meta, the original 
+creators of LLAMA, as well as the people behind the HuggingFace implementation of ALPACA