import gradio as gr
from transformers import TapexTokenizer, BartForConditionalGeneration
import pandas as pd

tokenizer = TapexTokenizer.from_pretrained("microsoft/tapex-large-finetuned-wtq")
model = BartForConditionalGeneration.from_pretrained("microsoft/tapex-large-finetuned-wtq")

data = {
    "year": [1896, 1900, 1904, 2004, 2008, 2012],
    "city": ["athens", "paris", "st. louis", "athens", "beijing", "london"]
}
table = pd.DataFrame.from_dict(data)

# tapex accepts uncased input since it is pre-trained on the uncased corpus
#query = "how many different countries had election in 21st century?"
#encoding = tokenizer(table=table, query=query, return_tensors="pt")

# outputs = model.generate(**encoding)

# print(tokenizer.batch_decode(outputs, skip_special_tokens=True))
# [' 2008.0']

def launch(input):
    encoding = tokenizer(table=table, query=input, return_tensors="pt")
    outputs=model.generate(**encoding)
    return tokenizer.batch_decode(outputs, skip_special_tokens=True)

iface = gr.Interface(launch,
                    inputs="text",
                    outputs="text")
iface.launch(share=True)