File size: 5,976 Bytes
dcc24f8 |
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 |
"""
Quantitative Evaluation for finance-lora-v6.
Tests the model on the full test dataset and computes accuracy metrics.
Author: Ranjit Behera
"""
import json
import subprocess
import sys
import re
from pathlib import Path
from collections import defaultdict
MODEL_PATH = "models/base/phi3-finance-base"
ADAPTER_PATH = "models/adapters/finance-lora-v6"
TEST_FILE = "data/synthetic/test_emails.json"
def generate(prompt: str) -> str:
"""Generate response using mlx_lm.generate."""
cmd = [
sys.executable, "-m", "mlx_lm.generate",
"--model", MODEL_PATH,
"--adapter-path", ADAPTER_PATH,
"--prompt", prompt,
"--max-tokens", "200"
]
try:
result = subprocess.run(cmd, capture_output=True, text=True, timeout=120)
return result.stdout
except Exception as e:
return f"Error: {e}"
def parse_json_from_output(output: str) -> dict:
"""Extract JSON from model output."""
try:
# Find JSON in output
match = re.search(r'\{[^{}]+\}', output, re.DOTALL)
if match:
return json.loads(match.group())
except:
pass
return {}
def compare_entities(expected: dict, predicted: dict) -> dict:
"""Compare expected vs predicted entities."""
fields = ['amount', 'type', 'date', 'account', 'reference', 'merchant']
results = {}
for field in fields:
exp_val = str(expected.get(field, '')).lower().strip()
pred_val = str(predicted.get(field, '')).lower().strip()
if exp_val:
# Normalize amount (remove commas, trailing zeros)
if field == 'amount':
exp_val = exp_val.replace(',', '').rstrip('0').rstrip('.')
pred_val = pred_val.replace(',', '').rstrip('0').rstrip('.')
results[field] = {
'expected': exp_val,
'predicted': pred_val,
'correct': exp_val == pred_val
}
return results
def run_evaluation(limit: int = None):
"""Run evaluation on test dataset."""
print("=" * 70)
print("📊 QUANTITATIVE EVALUATION - finance-lora-v6")
print("=" * 70)
print(f"Model: {MODEL_PATH}")
print(f"Adapter: {ADAPTER_PATH}")
print()
# Load test data
with open(TEST_FILE) as f:
test_data = json.load(f)
if limit:
test_data = test_data[:limit]
print(f"Testing {len(test_data)} samples...")
print()
# Track results by bank and field
bank_stats = defaultdict(lambda: {'correct': 0, 'total': 0})
field_stats = defaultdict(lambda: {'correct': 0, 'total': 0})
for i, sample in enumerate(test_data):
bank = sample.get('bank', 'unknown')
body = sample.get('body', sample.get('raw_text', ''))
expected = sample.get('entities', {})
# Create prompt
prompt = f"""Extract financial entities from this email:
{body}
Extract: amount, type, date, account, reference, merchant
Output JSON:"""
# Generate prediction
output = generate(prompt)
predicted = parse_json_from_output(output)
# Compare
comparison = compare_entities(expected, predicted)
# Update stats
sample_correct = 0
sample_total = 0
for field, result in comparison.items():
field_stats[field]['total'] += 1
if result['correct']:
field_stats[field]['correct'] += 1
sample_correct += 1
sample_total += 1
if sample_total > 0:
bank_stats[bank]['total'] += 1
if sample_correct == sample_total:
bank_stats[bank]['correct'] += 1
# Progress
if (i + 1) % 5 == 0:
print(f" Processed {i + 1}/{len(test_data)}...")
# Print results
print()
print("=" * 70)
print("📈 RESULTS BY BANK")
print("=" * 70)
total_correct = 0
total_samples = 0
for bank in sorted(bank_stats.keys()):
stats = bank_stats[bank]
acc = stats['correct'] / stats['total'] * 100 if stats['total'] > 0 else 0
status = "✅" if acc >= 90 else "⚠️" if acc >= 70 else "❌"
print(f" {bank.upper():12} {stats['correct']:3}/{stats['total']:3} = {acc:5.1f}% {status}")
total_correct += stats['correct']
total_samples += stats['total']
overall_acc = total_correct / total_samples * 100 if total_samples > 0 else 0
print(f"\n {'OVERALL':12} {total_correct:3}/{total_samples:3} = {overall_acc:5.1f}%")
print()
print("=" * 70)
print("📈 RESULTS BY FIELD")
print("=" * 70)
for field in ['amount', 'type', 'date', 'account', 'reference', 'merchant']:
if field in field_stats:
stats = field_stats[field]
acc = stats['correct'] / stats['total'] * 100 if stats['total'] > 0 else 0
status = "✅" if acc >= 90 else "⚠️" if acc >= 70 else "❌"
print(f" {field:12} {stats['correct']:3}/{stats['total']:3} = {acc:5.1f}% {status}")
print()
print("=" * 70)
print("✅ Evaluation Complete!")
print("=" * 70)
# Save results
results = {
'model': MODEL_PATH,
'adapter': ADAPTER_PATH,
'total_samples': total_samples,
'overall_accuracy': overall_acc,
'by_bank': {k: v for k, v in bank_stats.items()},
'by_field': {k: v for k, v in field_stats.items()}
}
with open('evaluation_results_v6.json', 'w') as f:
json.dump(results, f, indent=2)
print(f"\n💾 Results saved to evaluation_results_v6.json")
if __name__ == "__main__":
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('--limit', type=int, default=20, help='Number of samples to test')
args = parser.parse_args()
run_evaluation(limit=args.limit)
|