Spaces:
Running
Running
File size: 7,588 Bytes
25e624c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 |
# -*- coding: utf-8 -*-
"""
Comprehensive Emotion Test - Test recognition across many text strings
Tests:
1. Greetings (should be neutral)
2. Questions (should be neutral/curious)
3. Positive emotions
4. Negative emotions
5. Complex/mixed emotions
6. Edge cases
Ensures the system interprets text correctly.
"""
import sys
import os
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
from avatar.sentiment_transformer import SentimentAnalyzer as BinaryAnalyzer
from avatar.sentiment_multi_emotion import MultiEmotionAnalyzer
from avatar.sentiment_emoji_map import EmojiMapper
def run_comprehensive_test():
"""Run comprehensive emotion recognition test"""
print("=" * 80)
print("COMPREHENSIVE EMOTION RECOGNITION TEST")
print("=" * 80)
print()
# Load analyzers
print("Loading analyzers...")
binary = BinaryAnalyzer()
multi = MultiEmotionAnalyzer()
mapper = EmojiMapper()
# Test cases organized by category
test_cases = {
"GREETINGS (should be neutral/positive)": [
("How are you?", "neutral"),
("Hello! How are you?", "neutral"),
("Hey, how are you?", "neutral"),
("Hi there!", "neutral"),
("What's up?", "neutral"),
("Good morning!", "neutral"),
("Good evening!", "neutral"),
("How is it going?", "neutral"),
("How have you been?", "neutral"),
("Are you okay?", "neutral"),
("Yo!", "neutral"),
("Hey!", "neutral"),
("Greetings!", "neutral"),
],
"QUESTIONS (should be neutral/curious)": [
("What time is it?", "neutral"),
("Where is the store?", "neutral"),
("Can you help me?", "neutral"),
("Do you know the answer?", "neutral"),
("What do you think?", "neutral"),
("Is this correct?", "neutral"),
("Why did that happen?", "neutral"),
("How does this work?", "neutral"),
("What is the meaning of life?", "neutral"),
("Who is that?", "neutral"),
],
"POSITIVE EMOTIONS": [
("I am so happy today!", "positive"),
("I love this!", "positive"),
("This is amazing!", "positive"),
("Thank you so much!", "positive"),
("I'm grateful for everything", "positive"),
("This makes me so excited!", "positive"),
("I feel wonderful!", "positive"),
("You're the best!", "positive"),
("I'm thrilled about this!", "positive"),
("What a beautiful day!", "positive"),
("I can't wait!", "positive"),
("This is fantastic!", "positive"),
("I'm so proud of you!", "positive"),
("Best day ever!", "positive"),
("I'm feeling great!", "positive"),
],
"NEGATIVE EMOTIONS": [
("I am so sad", "negative"),
("This is terrible", "negative"),
("I'm really angry", "negative"),
("This makes me furious", "negative"),
("I hate this", "negative"),
("I'm scared", "negative"),
("This is awful", "negative"),
("I feel miserable", "negative"),
("I'm disappointed", "negative"),
("This is frustrating", "negative"),
("I can't stand this", "negative"),
("I'm heartbroken", "negative"),
("Everything is wrong", "negative"),
("I feel hopeless", "negative"),
("This is disgusting", "negative"),
],
"NEUTRAL STATEMENTS": [
("The weather is okay", "neutral"),
("It's Monday", "neutral"),
("The meeting is at 3pm", "neutral"),
("I need to go shopping", "neutral"),
("The car is blue", "neutral"),
("I'm going to work", "neutral"),
("It's raining outside", "neutral"),
("The document is ready", "neutral"),
("I'll call you later", "neutral"),
("The report is on my desk", "neutral"),
],
"EDGE CASES": [
("I'm not happy", "negative"),
("I don't feel bad", "positive"),
("This isn't terrible", "neutral"),
("Meh", "neutral"),
("Whatever", "neutral"),
("Fine", "neutral"),
("Ok", "neutral"),
("Sure", "neutral"),
("I guess", "neutral"),
("Maybe", "neutral"),
],
"SARCASM (tricky)": [
("Oh great, just what I needed", "negative"),
("Yeah, that's exactly what I wanted", "negative"),
("Oh wonderful", "negative"),
("How fantastic", "negative"),
("Just perfect", "negative"),
],
"MIXED EMOTIONS (should detect last sentiment)": [
("I was sad, but now I'm happy!", "positive"),
("Started great, ended terribly", "negative"),
("Good news and bad news", "neutral"),
("I love this! Wait, no I hate it", "negative"),
("Happy at first, now confused", "neutral"),
],
}
# Run tests
results = {"pass": 0, "fail": 0, "total": 0}
failed_tests = []
for category, tests in test_cases.items():
print(f"\n{'='*80}")
print(f"📋 {category}")
print(f"{'='*80}")
for text, expected_polarity in tests:
results["total"] += 1
# Test binary analyzer
binary_result = binary.analyze(text)
binary_label = binary_result["label"]
# Test multi-emotion analyzer
multi_result = multi.analyze(text)
multi_label = multi_result["label"]
multi_polarity = multi_result["polarity"]
# Get emoji
emoji = mapper.get_emoji(multi_label)
# Determine pass/fail based on polarity expectation
if expected_polarity == "neutral":
# Neutral can also be positive for greetings
passed = multi_polarity in ["neutral", "positive"]
elif expected_polarity == "positive":
passed = multi_polarity == "positive"
elif expected_polarity == "negative":
passed = multi_polarity == "negative"
else:
passed = True
status = "✅" if passed else "❌"
if passed:
results["pass"] += 1
else:
results["fail"] += 1
failed_tests.append((text, expected_polarity, multi_label, multi_polarity))
print(f"{status} '{text[:40]:40}' -> {multi_label:15} ({multi_polarity:8}) {emoji}")
# Summary
print("\n" + "=" * 80)
print("📊 SUMMARY")
print("=" * 80)
print(f"Total tests: {results['total']}")
print(f"Passed: {results['pass']} ({results['pass']/results['total']*100:.1f}%)")
print(f"Failed: {results['fail']} ({results['fail']/results['total']*100:.1f}%)")
if failed_tests:
print("\n❌ FAILED TESTS:")
print("-" * 80)
for text, expected, detected, polarity in failed_tests:
print(f" '{text[:50]}' - expected {expected}, got {detected} ({polarity})")
return results
if __name__ == "__main__":
run_comprehensive_test()
|