app.py

from transformers import AutoTokenizer, AutoModelForSequenceClassification
from scipy.special import expit
import numpy as np
import os
import gradio as gr
import requests

# set up model
authtoken = os.environ.get("TOKEN") or True
tokenizer = AutoTokenizer.from_pretrained("guidecare/feelings_and_issues", use_auth_token=authtoken)
model = AutoModelForSequenceClassification.from_pretrained("guidecare/feelings_and_issues", use_auth_token=authtoken)
all_label_names = list(model.config.id2label.values())


def predict(text):
    probs = expit(model(**tokenizer([text], return_tensors="pt", padding=True)).logits.detach().numpy())
    # can't use numpy for whatever reason
    probs = [float(np.round(i, 2)) for i in probs[0]]
    # break out issue, harm, sentiment, feeling
    zipped_list = list(zip(all_label_names, probs))
    print(text, zipped_list)
    issues = [(i, j) for i, j in zipped_list if i.startswith('issue')]
    feelings = [(i, j) for i, j in zipped_list if i.startswith('feeling')]
    harm = [(i, j) for i, j in zipped_list if i.startswith('harm')]
    # keep tops for each one
    issues = sorted(issues, key=lambda x: x[1])[::-1][:3]
    feelings = sorted(feelings, key=lambda x: x[1])[::-1][:3]
    harm = sorted(harm, key=lambda x: x[1])[::-1][:1]
    # top is the combo of these
    top = issues + feelings + harm

    logToNotion(text, top)
    
    d = {i: j for i, j in top}
    return d

def logToNotion(text, top):
    url = "https://api.notion.com/v1/pages"

    payload = {
        "properties": {
            "title": {
                "title": [{
                    "text": {
                        "content": "."
                    }
                }]
            },
            "input": {
                "rich_text": [{
                    "text": {
                        "content": text
                    }
                }]
            },
            "output": {
                "rich_text": [{
                    "text": {
                        "content": ", ".join(str(x) for x in top)
                    }
                }]
            }
        },
        "parent": "4a220773ac694851811e87f4571ec41d"
    }
    headers = {
        "Accept": "application/json",
        "Notion-Version": "2022-02-22",
        "Content-Type": "application/json",
        "Authorization": "Bearer secret_lc3Weq9s87EnkNWk2Yibko91hhLuNQIEzSoeAAoriUz"
    }
    
    response = requests.post(url, json=payload, headers=headers)

iface = gr.Interface(
  fn=predict, 
  inputs="text",
  outputs="label",
  #examples=["This test tomorrow is really freaking me out."]
)

iface.launch()