Views
No views yet
1from transformers import RobertaForSequenceClassification, RobertaTokenizer
2import torch
3
4output_model_dir = 'RappyProgramming/IPW-RoBERTa-uncased'
5model = RobertaForSequenceClassification.from_pretrained(output_model_dir)
6tokenizer = RobertaTokenizer.from_pretrained(output_model_dir)
71input_texts = [
2 "this meeting is scheduled for next week",
3 "drop dead",
4 "you're the best friend i could ever have in this whole wide world!!"
5]1inputs = tokenizer(input_texts, return_tensors="pt", padding=True, truncation=True)
2with torch.no_grad():
3 outputs = model(**inputs)
4
5predicted_class_indices = torch.argmax(outputs.logits, dim=1).tolist()
6probs = torch.softmax(outputs.logits, dim=1).tolist()
7labels = ["Negative", "Neutral", "Positive"]
8
9for i, input_text in enumerate(input_texts):
10 predicted_label = labels[predicted_class_indices[i]]
11 predicted_probabilities = {label: prob for label, prob in zip(labels, probs[i])}
12
13 print(f"Input text {i+1}: {input_text}")
14 print(f"Predicted label: {predicted_label}")
15 print("Predicted probabilities:")
16
17 for label, prob in predicted_probabilities.items():
18 print(f"{label}: {prob:.4f}")
19
20 print()Output:
Input text 1: this meeting is scheduled for next week
Predicted label: Neutral
Predicted probabilities:
Negative: 0.0002
Neutral: 0.9985
Positive: 0.0012
Input text 2: drop dead
Predicted label: Negative
Predicted probabilities:
Negative: 0.5771
Neutral: 0.4225
Positive: 0.0005
Input text 3: you're the best friend i could ever have in this whole wide world!!
Predicted label: Positive
Predicted probabilities:
Negative: 0.0003
Neutral: 0.0001
Positive: 0.9996