MultiModal-Phi2 / README.md
ravi.naik
Fixed relative import issues
667ae00
---
title: MultiModal Phi2
emoji: πŸš€
colorFrom: blue
colorTo: red
sdk: gradio
sdk_version: 3.35.2
app_file: app.py
pinned: false
license: mit
---
## Phi2 : Multimodal Finetuning
### Details
1. LLM Backbone: Phi2
2. Vision Tower: clip-vit-large-patch14-336
3. Audio Model: Whisper
4. Pretraining Dataset: LAION-CC-SBU dataset with BLIP captions(200k samples)
5. Finetuning Dataset: Instruct 150k dataset based on COCO
### Design
![image](https://github.com/RaviNaik/ERA-CAPSTONE/assets/23289802/56df24cd-2681-4e17-ab64-9652f609b15f)
### Pretraining
#### Training Loss Curve
![image](https://github.com/RaviNaik/ERA-CAPSTONE/assets/23289802/b6c37a95-0a56-4b52-8719-3ff56dc1b703)
#### Learing Rate
![image](https://github.com/RaviNaik/ERA-CAPSTONE/assets/23289802/44d9a11b-b28d-47e1-ba1d-d6dc22ebe748)
#### Training Logs
![image](https://github.com/RaviNaik/ERA-CAPSTONE/assets/23289802/76543d98-d9fe-4c1a-ac47-3d06e48053ad)
### Finetuning
#### Training Loss Curve
![image](https://github.com/RaviNaik/ERA-CAPSTONE/assets/23289802/45ef40bd-fae5-4cfe-a522-c0eed2833230)
#### Learing Rate
![image](https://github.com/RaviNaik/ERA-CAPSTONE/assets/23289802/df60ee62-a537-4e36-a7f7-f7111e101162)
#### Training Logs
![image](https://github.com/RaviNaik/ERA-CAPSTONE/assets/23289802/2747acce-bc99-4c37-a05a-d5e81cb9aa9d)
### Results
![image](https://github.com/RaviNaik/ERA-CAPSTONE/assets/23289802/f12a9f04-df32-413e-b957-774c30381b2b)