added anther model

app.py

@@ -1,35 +1,19 @@
 import gradio as gr
-import torch
-import tensorflow as tf
-from transformers import AutoTokenizer
-from model import SentimentClassifier
+from fastai.vision.all import *
 
-model_state_dict = tf.keras.load_model('sentimentality.h5')
-model = SentimentClassifier(2)
-model.load_state_dict(model_state_dict)
-model.eval()
+# Load the model
+learn = load_learner('sentimentality.h5')
 
-tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')
+# Prediction function
+def make_prediction(user_sentence):
+  
+  prediction = learn.predict(user_sentence)
+  dict = {'1': 'Negative', '2': 'Neutral', '3': 'Positive'}
+  return dict[prediction[0]]
 
+title = "Sentiment Analysis MyAnimeList Reviews with fastai"
+description = "<p style='text-align: center'>Identifier si un commentaire dans MyAnimeList est positif, neutre ou négatif.<br/> Permet de connaître rapidement le sentiment globale que dégage un avis sur le site.</p>"
 
-def preprocess(text):
-    inputs = tokenizer(text, padding='max_length',
-                       truncation=True, max_length=512, return_tensors='pt')
-    return inputs
-# Define a function to use the model to make predictions
-def predict(review):
-    inputs = preprocess(review)
-    with torch.no_grad():
-        outputs = model(inputs['input_ids'], inputs['attention_mask'])
-        predicted_class = torch.argmax(outputs[0]).item()
-    if(predicted_class==0):
-        return "It was a negative review"
-    return "It was a positive review"
+app = gr.Interface(fn=make_prediction, title=title, description=description, inputs=gr.TextArea(), outputs='text')
 
-# Create a Gradio interface
-input_text = gr.inputs.Textbox(label="Input Text")
-output_text = gr.outputs.Textbox(label="Output Text")
-interface = gr.Interface(fn=predict, inputs=input_text, outputs=output_text)
-
-# Run the interface
-interface.launch()
+app.launch()