# #Import the required Libraries
# import gradio as gr
# import pickle
# import pandas as pd
# import numpy as np
# import transformers

# # Load from the Hugging Face model hub
# model = transformers.AutoModel.from_pretrained('AmpomahChief/sentiment_analysis_on_covid_tweets')


# # Define a function that takes in input and passes it through the model
# def predict(inputs):
#     input_ids = transformers.BertTokenizer.from_pretrained('AmpomahChief/sentiment_analysis_on_covid_tweets').encode(inputs, return_tensors='pt')
#     output = model(input_ids)[0]
#     return output

# # Create a Gradio interface for the model
# interface = gr.Interface(fn=predict, inputs=gr.Textbox(prompt="Input text:"), outputs=gr.Textbox(prompt="Model output:"))

# # Launch the interface
# interface.launch()



import gradio as gr
# Creating a gradio app using the inferene API
App = gr.Interface.load("huggingface/AmpomahChief/sentiment_analysis_on_covid_tweets",
  title="COVID 19 tweets sentiment analysis", description ="This is a sentiment analysis on COVID 19 tweets using pretrained model on hugging face",
 allow_flagging=False, examples=[["Input your text here"]]
)

App.launch()





# import gradio as gr
# from transformers import pipeline
# import transformers

# Model = transformers.AutoModel.from_pretrained('AmpomahChief/sentiment_analysis_on_covid_tweets')
# pipeline = pipeline(task="image-classification", model=Model)

# def predict(inputs):
#     input_ids = transformers.BertTokenizer.from_pretrained('AmpomahChief/sentiment_analysis_on_covid_tweets').encode(inputs, return_tensors='pt')
#     output = model(input_ids)[0]
#     return output

# # Create a Gradio interface for the model
# interface = gr.Interface(fn=predict, inputs=gr.Textbox(prompt="Input text:"), outputs=gr.Textbox(prompt="Model output:"))

# # Launch the interface
# interface.launch()