necpfzlakay/com.vqa_backend

Back-end Python server of the Visual Question Answering as Microservices final project

com.VQA_Backend

Back-end Python server of the Visual Question Answering as Microservices final project

How To Use

• Download the huggingface model via url given below:

https://cdn-lfs.huggingface.co/dandelin/vilt-b32-finetuned-vqa/4d5f3409947b0369487ece7c5868f0040ceb67d25735dbb4ac5e99e03bab3a19

• then paste it to the model folder and change name of model to 'pytorch_model.bin'
• After that, Download the main modal by given link below, and paste it to same folder with ApiService.py file

https://drive.google.com/file/d/16UQByRv_jzrlZ0IiyqkDzB_6YdJ_2wWA/view?usp=sharing

• and just double click ApiService.py file
• server will run on the http://31.145.7.41/ url
• Note: Our server have 31.145.7.41 external ip, so if your server have not this external ip server will work on just local network

Needed Dependency Import Libraries

• To run this import within an order (Just run on terminal them if you did not install before)

pip3 install flask
pip3 install torch torchvision torchaudio
pip3 install werkzeug
pip3 install transformers
pip3 install pillow
pip3 install BytesIO
pip3 install pyodbc
pip3 install json
pip3 install pymssql
pip3 install requests