Upload application.py

application.py

@@ -0,0 +1,120 @@
+# -*- coding: utf-8 -*-
+"""Application.ipynb
+
+Automatically generated by Colaboratory.
+
+Original file is located at
+    https://colab.research.google.com/drive/148du8431_JkTaH-totdocC2aUXzOWimL
+"""
+
+from google.colab import drive
+drive.mount('/content/drive')
+
+pip install transformers
+
+pip install gradio
+
+from transformers import BertTokenizer, TFBertForSequenceClassification
+import tensorflow as tf
+
+# Load tokenizer
+tokenizer = BertTokenizer.from_pretrained("nlpaueb/bert-base-greek-uncased-v1")
+
+# Load model
+model = TFBertForSequenceClassification.from_pretrained('/content/drive/MyDrive/Mini_Project/new_emdedding trial')
+
+def check_sarcasm(sentence):
+    tf_batch = tokenizer(sentence, max_length=128, padding=True, truncation=True, return_tensors='tf')
+    tf_outputs = model(tf_batch.input_ids, tf_batch.token_type_ids)
+    tf_predictions = tf.nn.softmax(tf_outputs.logits, axis=-1)
+    pred_label = tf.argmax(tf_predictions, axis=1)
+
+    if pred_label == 1:
+        return "Sarcastic"
+    else:
+        return "Not sarcastic"
+
+# Example usage
+sentence = "Μεξικό: 25 νεκροί από την πτώση λεωφορείου στον γκρεμό"
+result = check_sarcasm(sentence)
+print(result)
+
+import gradio as gr
+
+def check_sarcasm(sentence):
+    tf_batch = tokenizer(sentence, max_length=128, padding=True, truncation=True, return_tensors='tf')
+    tf_outputs = model(tf_batch.input_ids, tf_batch.token_type_ids)
+    tf_predictions = tf.nn.softmax(tf_outputs.logits, axis=-1)
+    pred_label = tf.argmax(tf_predictions, axis=1)
+
+    if pred_label == 1:
+        return "Sarcastic"
+    else:
+        return "Not sarcastic"
+
+# Create a Gradio interface
+iface = gr.Interface(
+    fn=check_sarcasm,
+    inputs="text",
+    outputs="text",
+    title="Sarcasm Detection",
+    description="Enter a headline and check if it's sarcastic."
+)
+
+# Launch the interface
+iface.launch(share=True)
+
+from transformers import pipeline
+import gradio as gr
+
+model = pipeline("Sarcasm Detection" , model="/content/drive/MyDrive/Mini_Project/new_emdedding trial/tf_model.h5")
+def check_sarcasm(sentence):
+    tf_batch = tokenizer(sentence, max_length=128, padding=True, truncation=True, return_tensors='tf')
+    tf_outputs = model(tf_batch.input_ids, tf_batch.token_type_ids)
+    tf_predictions = tf.nn.softmax(tf_outputs.logits, axis=-1)
+    pred_label = tf.argmax(tf_predictions, axis=1)
+
+    if pred_label == 1:
+        return "Sarcastic"
+    else:
+        return "Not sarcastic"
+# Create a Gradio interface
+iface = gr.Interface(
+    fn=check_sarcasm,
+    inputs="text",
+    outputs="text",
+    title="Sarcasm Detection",
+    description="Enter a headline and check if it's sarcastic."
+)
+
+# Launch the interface
+iface.launch()
+
+import gradio as gr
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+
+tokenizer = AutoTokenizer.from_pretrained("nlpaueb/bert-base-greek-uncased-v1")
+model = AutoModelForSequenceClassification.from_pretrained("/content/drive/MyDrive/Mini_Project/new_emdedding trial",from_tf=True)
+def check_sarcasm(sentence):
+    tf_batch = tokenizer(sentence, max_length=128, padding=True, truncation=True, return_tensors='tf')
+    tf_outputs = model(tf_batch.input_ids, tf_batch.token_type_ids)
+    tf_predictions = tf.nn.softmax(tf_outputs.logits, axis=-1)
+    pred_label = tf.argmax(tf_predictions, axis=1)
+
+    if pred_label == 1:
+        return "Sarcastic"
+    else:
+        return "Not sarcastic"
+
+# Create a Gradio interface
+iface = gr.Interface(
+    fn=check_sarcasm,
+    inputs="text",
+    outputs="text",
+    title="Sarcasm Detection",
+    description="Enter a headline and check if it's sarcastic."
+)
+
+# Launch the interface
+iface.launch(share=True)
+