Spaces:
Sleeping
Sleeping
Uploaded required files
Browse files- app.py +60 -0
- requirements.txt +4 -0
app.py
ADDED
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# -*- coding: utf-8 -*-
|
2 |
+
"""Sentiment Analysis App.ipynb
|
3 |
+
|
4 |
+
Automatically generated by Colaboratory.
|
5 |
+
|
6 |
+
Original file is located at
|
7 |
+
https://colab.research.google.com/drive/1t6wAnMPDdEHuioRZofR8_JEPrzuT7KAJ
|
8 |
+
"""
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
# Import the required Libraries
|
13 |
+
import gradio as gr
|
14 |
+
import numpy as np
|
15 |
+
import transformers
|
16 |
+
from transformers import AutoTokenizer, AutoConfig, AutoModelForSequenceClassification, TFAutoModelForSequenceClassification
|
17 |
+
from scipy.special import softmax
|
18 |
+
|
19 |
+
|
20 |
+
# Requirements
|
21 |
+
model_path = "Queensly/finetuned_albert_base_v2"
|
22 |
+
tokenizer = AutoTokenizer.from_pretrained(model_path)
|
23 |
+
config = AutoConfig.from_pretrained(model_path)
|
24 |
+
model = AutoModelForSequenceClassification.from_pretrained(model_path)
|
25 |
+
|
26 |
+
# Preprocess text (username and link placeholders)
|
27 |
+
def preprocess(text):
|
28 |
+
new_text = []
|
29 |
+
for t in text.split(" "):
|
30 |
+
t = "@user" if t.startswith("@") and len(t) > 1 else t
|
31 |
+
t = "http" if t.startswith("http") else t
|
32 |
+
new_text.append(t)
|
33 |
+
return " ".join(new_text)
|
34 |
+
|
35 |
+
#Function to process the input and return prediction
|
36 |
+
def sentiment_analysis(text):
|
37 |
+
text = preprocess(text)
|
38 |
+
|
39 |
+
encoded_input = tokenizer(text, return_tensors = "pt") # for PyTorch-based models
|
40 |
+
output = model(**encoded_input)
|
41 |
+
scores_ = output[0][0].detach().numpy()
|
42 |
+
scores_ = softmax(scores_)
|
43 |
+
|
44 |
+
#Output of scores by converting a list of labels and scores into a dictionary format
|
45 |
+
labels = ["Negative", "Neutral", "Positive"]
|
46 |
+
scores = {l:float(s) for (l,s) in zip(labels, scores_) }
|
47 |
+
return scores
|
48 |
+
|
49 |
+
|
50 |
+
#App interface with gradio
|
51 |
+
app = gr.Interface(fn = sentiment_analysis,
|
52 |
+
inputs = gr.Textbox("Write your text or tweet here..."),
|
53 |
+
outputs = "label",
|
54 |
+
title = "Sentiment Analysis of Tweets on COVID-19 Vaccines",
|
55 |
+
description = "This app analyzes sentiment of text based on tweets about COVID-19 Vaccines using a fine-tuned albert_base_v2 model",
|
56 |
+
interpretation = "default",
|
57 |
+
examples=[["covid vaccines are great!"]]
|
58 |
+
)
|
59 |
+
|
60 |
+
app.launch()
|
requirements.txt
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
numpy
|
2 |
+
scipy
|
3 |
+
torch
|
4 |
+
transformers
|