YureiYuri commited on
Commit
223a572
·
verified ·
1 Parent(s): d72b347

update README

Browse files
Files changed (1) hide show
  1. README.md +71 -4
README.md CHANGED
@@ -1,8 +1,75 @@
1
  ---
2
- license: mit
3
  language:
4
  - en
5
- pipeline_tag: table-question-answering
6
  tags:
7
- - mental_health
8
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
 
2
  language:
3
  - en
4
+ license: apache-2.0
5
  tags:
6
+ - mental-health
7
+ - therapy
8
+ - chatbot
9
+ - text-classification
10
+ - distilbert
11
+ pipeline_tag: text-classification
12
+ ---
13
+
14
+ # Therapy Chatbot - Mental Health Intent Classification
15
+
16
+ This model classifies mental health-related text into 11 different intent categories.
17
+
18
+ ## Model Description
19
+
20
+ - **Model type:** DistilBERT for Sequence Classification
21
+ - **Language:** English
22
+ - **Training data:** 3,508 therapy context-response pairs
23
+ - **Task:** Intent classification for mental health chatbot
24
+
25
+ ## Intent Categories
26
+
27
+ The model classifies text into these 11 intents:
28
+ - `depression` - Feelings of sadness, hopelessness, worthlessness
29
+ - `anxiety` - Worry, panic, nervousness
30
+ - `self_esteem` - Issues with self-worth and confidence
31
+ - `relationship` - Relationship problems and conflicts
32
+ - `family` - Family-related concerns
33
+ - `sleep_issues` - Insomnia and sleep problems
34
+ - `anger` - Anger management issues
35
+ - `suicide` - Suicidal thoughts (crisis detection)
36
+ - `trauma` - PTSD and trauma-related issues
37
+ - `grief` - Loss and bereavement
38
+ - `general_support` - General emotional support needs
39
+
40
+ ## Usage
41
+ ```python
42
+ from transformers import DistilBertTokenizerFast, DistilBertForSequenceClassification
43
+ import torch
44
+
45
+ # Load model and tokenizer
46
+ model = DistilBertForSequenceClassification.from_pretrained("YOUR-USERNAME/therapy-chatbot")
47
+ tokenizer = DistilBertTokenizerFast.from_pretrained("YOUR-USERNAME/therapy-chatbot")
48
+
49
+ # Make prediction
50
+ text = "I feel worthless and can't sleep at night"
51
+ inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
52
+
53
+ with torch.no_grad():
54
+ outputs = model(**inputs)
55
+
56
+ prediction = torch.argmax(outputs.logits, dim=1).item()
57
+ print(f"Predicted intent: {prediction}")
58
+ ```
59
+
60
+ ## Training
61
+
62
+ - **Epochs:** 6
63
+ - **Batch size:** 16
64
+ - **Learning rate:** 3e-5
65
+ - **Validation loss:** ~0.030
66
+
67
+ ## Limitations
68
+
69
+ ⚠️ **Important:** This model is NOT a replacement for professional mental health care. It's designed to assist in categorizing user messages for appropriate responses.
70
+
71
+ ## Disclaimer
72
+
73
+ This is an AI model for educational and support purposes only. In case of mental health emergencies, please contact:
74
+ - National Suicide Prevention Lifeline: 988
75
+ - Crisis Text Line: Text HOME to 741741