ShanukaB commited on
Commit
e6be4c7
·
verified ·
1 Parent(s): 3dd5123

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +40 -4
README.md CHANGED
@@ -1,7 +1,7 @@
1
  ---
2
- title: Emotion Recognition (English)
3
  emoji: 😊
4
- colorFrom: blue
5
  colorTo: purple
6
  sdk: gradio
7
  sdk_version: 4.44.0
@@ -10,5 +10,41 @@ python_version: 3.11
10
  pinned: false
11
  ---
12
 
13
- Simple English tweet emotion classifier using TF-IDF + Logistic Regression
14
- Model repo: E-motionAssistant/Englsih_Trained_Model_LR
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ title: Multilingual Emotion Recognition (EN / SI / TA)
3
  emoji: 😊
4
+ colorFrom: indigo
5
  colorTo: purple
6
  sdk: gradio
7
  sdk_version: 4.44.0
 
10
  pinned: false
11
  ---
12
 
13
+ # Multilingual Text Emotion Recognition App
14
+ **English සිංහල • தமிழ்**
15
+
16
+ This is a simple web app that detects emotions from text in three languages:
17
+
18
+ - **English** → Logistic Regression + TF-IDF (classical ML model)
19
+ - **Sinhala** → Finetuned transformer model
20
+ - **Tamil** → Finetuned transformer model
21
+
22
+ ### Supported Languages & Models
23
+ | Language | Model Type | Repository on Hugging Face |
24
+ |------------|-------------------------|------------------------------------------------------------------|
25
+ | English | TF-IDF + Logistic Regression | E-motionAssistant/Englsih_Trained_Model_LR |
26
+ | Sinhala | Finetuned Transformer | E-motionAssistant/SInhala_Text_Emotion_Recognition_Model |
27
+ | Tamil | Finetuned Transformer | E-motionAssistant/Tamil_Emotion_Recognition_Model |
28
+
29
+ ### How to Use
30
+ 1. Go to one of the tabs (English / සිංහල / தமிழ்)
31
+ 2. Type or paste your text/message/tweet
32
+ 3. The app will show the predicted emotion + confidence score
33
+
34
+ ### Examples
35
+ - English: "I hate mondays so much" → likely **sadness** or **anger**
36
+ - Sinhala: "මම ගොඩක් දුකින් ඉන්නවා" → likely **sadness**
37
+ - Tamil: "எனக்கு ரொம்ப கோபமா இருக்கு" → likely **anger**
38
+
39
+ ### Technical Notes
40
+ - Built with **Gradio** (v4.44.0)
41
+ - Models are downloaded from Hugging Face at startup (using `hf_hub_download` and `pipeline`)
42
+ - Uses **Python 3.11** to avoid compatibility issues
43
+ - No GPU required — runs on CPU
44
+
45
+ ### Want to run locally?
46
+ ```bash
47
+ git clone https://huggingface.co/spaces/YOUR_USERNAME/Space4
48
+ cd Space4
49
+ pip install -r requirements.txt
50
+ python app.py