0xnu commited on
Commit
8c084ea
1 Parent(s): 9cb476d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +38 -0
README.md CHANGED
@@ -38,3 +38,41 @@ Notes:
38
  - The `tokenize_chinese_chars` parameter indicates special handling for Chinese characters.
39
  - The maximum model length is set to 512 tokens.
40
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
38
  - The `tokenize_chinese_chars` parameter indicates special handling for Chinese characters.
39
  - The maximum model length is set to 512 tokens.
40
 
41
+ ## Run the model
42
+
43
+ ```Python
44
+ from transformers import AutoTokenizer, AutoModelForSequenceClassification
45
+
46
+ model_id = "0xnu/AGTD-v0.1"
47
+
48
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
49
+ model = AutoModelForSequenceClassification.from_pretrained(model_id)
50
+
51
+ # Input text
52
+ text = "This model trains on a diverse dataset and serves functions in applications requiring a mechanism for distinguishing between human and AI-generated text."
53
+
54
+ # Preprocess the text
55
+ inputs = tokenizer(text, return_tensors='pt')
56
+
57
+ # Run the model
58
+ outputs = model(**inputs)
59
+
60
+ # Interpret the output
61
+ logits = outputs.logits
62
+
63
+ # Apply softmax to convert logits to probabilities
64
+ probabilities = torch.softmax(logits, dim=1)
65
+
66
+ # Assuming the first class is 'human' and the second class is 'ai'
67
+ human_prob, ai_prob = probabilities.detach().numpy()[0]
68
+
69
+ # Print probabilities
70
+ print(f"Human Probability: {human_prob:.4f}")
71
+ print(f"AI Probability: {ai_prob:.4f}")
72
+
73
+ # Determine if the text is human or AI-generated
74
+ if human_prob > ai_prob:
75
+ print("The text is likely human-generated.")
76
+ else:
77
+ print("The text is likely AI-generated.")
78
+ ```