File size: 19,999 Bytes
60b21d3 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 | #!/usr/bin/env python3
# SPDX-FileCopyrightText: 2025 Stanford University, ETH Zurich, and the project authors (see CONTRIBUTORS.md)
# SPDX-FileCopyrightText: 2025 This source file is part of the OpenTSLM open-source project.
#
# SPDX-License-Identifier: MIT
"""
Test script for the ECG-QA CoT loader and dataset.
"""
import unittest
from opentslm.time_series_datasets.ecg_qa.ECGQACoTQADataset import ECGQACoTQADataset
from opentslm.logger import get_logger, set_global_verbose
def pretty_print_label_distribution(dataset, name):
"""Pretty print label distribution for a dataset."""
from opentslm.time_series_datasets.ecg_qa.ecgqa_cot_loader import get_label_distribution
label_dist = get_label_distribution(dataset)
total = len(dataset)
print(f"\n{name} dataset:")
print(f" Total samples: {total}")
print(f" Label distribution:")
for label, count in sorted(label_dist.items()):
print(f" {label:15s}: {count:5d} ({count/total*100:5.1f}%)")
class TestECGQACotLoader(unittest.TestCase):
"""
Unit tests for the ECG-QA CoT loader functions.
"""
def setUp(self):
# Set up global logger with verbose mode for detailed output
set_global_verbose(True)
self.logger = get_logger()
from opentslm.time_series_datasets.ecg_qa.ecgqa_cot_loader import load_ecg_qa_cot_splits
self.load_ecg_qa_cot_splits = load_ecg_qa_cot_splits
self.logger.loading("Loading ECG-QA CoT dataset splits...")
self.train, self.val, self.test = self.load_ecg_qa_cot_splits()
self.logger.success(f"Dataset loaded successfully: Train={len(self.train)}, Val={len(self.val)}, Test={len(self.test)}")
def test_dataset_sizes(self):
"""Test that the datasets are non-empty and splits are correct."""
self.logger.info("Testing dataset sizes...")
self.assertGreater(len(self.train), 0)
self.assertGreater(len(self.val), 0)
self.assertGreater(len(self.test), 0)
self.logger.success("Dataset size tests passed")
def test_label_distribution(self):
"""Test that label distributions are reasonable across splits."""
self.logger.info("Testing label distributions...")
pretty_print_label_distribution(self.train, "Train")
pretty_print_label_distribution(self.val, "Validation")
pretty_print_label_distribution(self.test, "Test")
from opentslm.time_series_datasets.ecg_qa.ecgqa_cot_loader import get_label_distribution
train_dist = get_label_distribution(self.train)
val_dist = get_label_distribution(self.val)
test_dist = get_label_distribution(self.test)
# Check that all splits have reasonable distributions
# Note: Not all labels need to appear in all splits due to random sampling
self.assertGreater(len(train_dist), 0)
self.assertGreater(len(val_dist), 0)
self.assertGreater(len(test_dist), 0)
# Check that major labels appear in all splits
major_labels = ['yes', 'no', 'none']
for label in major_labels:
if label in train_dist:
self.assertIn(label, val_dist, f"Major label '{label}' missing from validation set")
self.assertIn(label, test_dist, f"Major label '{label}' missing from test set")
self.logger.success("Label distribution tests passed")
def test_sample_keys(self):
"""Test that a sample contains all required keys."""
self.logger.info("Testing sample keys...")
sample = self.train[0]
required_keys = {"question", "question_type", "template_id", "ecg_id", "ecg_paths", "clinical_contexts", "rationale"}
self.assertTrue(required_keys.issubset(sample.keys()))
# Test that required fields are not None
for key in required_keys:
self.assertIsNotNone(sample[key], f"Required field '{key}' is None")
self.logger.success("Sample keys test passed")
def test_ecg_data_content(self):
"""Test that the ECG data is present and valid."""
self.logger.info("Testing ECG data content...")
sample = self.train[0]
# Check ECG IDs
self.assertIsInstance(sample["ecg_id"], list)
self.assertGreater(len(sample["ecg_id"]), 0)
for ecg_id in sample["ecg_id"]:
self.assertIsInstance(ecg_id, int)
# Check ECG paths
self.assertIsInstance(sample["ecg_paths"], list)
self.assertGreater(len(sample["ecg_paths"]), 0)
for path in sample["ecg_paths"]:
self.assertIsInstance(path, str)
self.assertTrue(path.endswith('.dat'))
# Check clinical contexts
self.assertIsInstance(sample["clinical_contexts"], list)
self.assertGreater(len(sample["clinical_contexts"]), 0)
for context in sample["clinical_contexts"]:
self.assertIsInstance(context, str)
self.assertGreater(len(context), 0)
self.logger.success("ECG data content tests passed")
def test_question_content(self):
"""Test that the question data is valid."""
self.logger.info("Testing question content...")
sample = self.train[0]
# Check question
self.assertIsInstance(sample["question"], str)
self.assertGreater(len(sample["question"]), 0)
# Check question type
self.assertIsInstance(sample["question_type"], str)
self.assertGreater(len(sample["question_type"]), 0)
# Check template ID
self.assertIsInstance(sample["template_id"], int)
self.logger.success("Question content tests passed")
def test_rationale_content(self):
"""Test that the rationale is a string and non-empty."""
self.logger.info("Testing rationale content...")
sample = self.train[0]
self.assertIsInstance(sample["rationale"], str)
self.assertGreater(len(sample["rationale"]), 0)
self.logger.success(f"Rationale test passed: length={len(sample['rationale'])}")
def test_cot_fields(self):
"""Test that CoT-specific fields are present."""
self.logger.info("Testing CoT-specific fields...")
sample = self.train[0]
# Check CoT fields that actually exist in the implementation
cot_fields = ["rationale", "template_id", "question_type"]
for field in cot_fields:
self.assertIn(field, sample)
# These should not be None
self.assertIsNotNone(sample[field], f"Required field '{field}' is None")
if field == "rationale":
self.assertIsInstance(sample[field], str)
elif field == "template_id":
self.assertIsInstance(sample[field], int)
elif field == "question_type":
self.assertIsInstance(sample[field], str)
self.logger.success("CoT fields tests passed")
def test_example_data(self):
"""Print example data to show what the dataset looks like."""
sample = self.train[0]
self.logger.info("="*80)
self.logger.info("EXAMPLE ECG-QA COT DATASET SAMPLE")
self.logger.info("="*80)
self.logger.info(f"Question: '{sample['question']}'")
self.logger.info(f"Question type: '{sample['question_type']}'")
self.logger.info(f"Template ID: {sample['template_id']}")
self.logger.info(f"ECG IDs: {sample['ecg_id']}")
self.logger.info(f"ECG paths: {sample['ecg_paths']}")
if 'rationale' in sample:
rationale_preview = sample['rationale'][:200] + "..." if len(sample['rationale']) > 200 else sample['rationale']
self.logger.info(f"Rationale: '{rationale_preview}'")
self.logger.info(f"Clinical contexts: {sample['clinical_contexts']}")
self.logger.info("="*80)
class TestECGQACoTQADataset(unittest.TestCase):
"""
Unit tests for the ECGQACoTQADataset class.
"""
def setUp(self):
# Set up global logger with verbose mode for detailed output
set_global_verbose(True)
self.logger = get_logger()
from opentslm.time_series_datasets.ecg_qa import ECGQACoTQADataset
self.ECGQACoTQADataset = ECGQACoTQADataset
self.logger.loading("Initializing ECGQACoTQADataset...")
# Use limited samples for faster testing
self.train_dataset = self.ECGQACoTQADataset(split="train", EOS_TOKEN="", max_samples=5)
self.val_dataset = self.ECGQACoTQADataset(split="validation", EOS_TOKEN="", max_samples=5)
self.test_dataset = self.ECGQACoTQADataset(split="test", EOS_TOKEN="", max_samples=5)
self.logger.success(f"Datasets initialized: Train={len(self.train_dataset)}, Val={len(self.val_dataset)}, Test={len(self.test_dataset)}")
# Test the exclude_comparison functionality
self.logger.loading("Testing exclude_comparison functionality...")
self.train_no_comparison = self.ECGQACoTQADataset(split="train", EOS_TOKEN="", max_samples=10, exclude_comparison=True)
self.logger.success(f"Non-comparison dataset initialized: {len(self.train_no_comparison)} samples")
def test_dataset_sizes(self):
"""Test that the datasets are non-empty and splits are correct."""
self.logger.info("Testing QA dataset sizes...")
self.assertGreater(len(self.train_dataset), 0)
self.assertGreater(len(self.val_dataset), 0)
self.assertGreater(len(self.test_dataset), 0)
self.logger.success("QA dataset size tests passed")
def test_sample_keys(self):
"""Test that a sample contains all required keys."""
self.logger.info("Testing QA sample keys...")
sample = self.train_dataset[0]
required_keys = {"answer", "pre_prompt", "post_prompt", "time_series", "time_series_text", "question", "question_type", "template_id"}
self.assertTrue(required_keys.issubset(sample.keys()))
# Test that required fields are not None
for key in required_keys:
self.assertIsNotNone(sample[key], f"Required field '{key}' is None")
self.logger.success("QA sample keys test passed")
def test_answer_is_rationale(self):
"""Test that the answer is a string (rationale)."""
self.logger.info("Testing answer format...")
sample = self.train_dataset[0]
self.assertIsInstance(sample["answer"], str)
self.assertGreater(len(sample["answer"]), 0)
self.logger.success(f"Answer test passed: length={len(sample['answer'])}")
def test_time_series_content(self):
"""Test that the time series and text are present and valid."""
self.logger.info("Testing time series content...")
sample = self.train_dataset[0]
self.assertIsInstance(sample["time_series"], list)
self.assertIsInstance(sample["time_series_text"], list)
# Should have multiple time series (one per ECG lead)
self.assertGreater(len(sample["time_series"]), 0)
self.assertGreater(len(sample["time_series_text"]), 0)
# Each time series should be non-empty
for i, ts in enumerate(sample["time_series"]):
self.assertGreater(len(ts), 0, f"Time series {i} is empty")
# Check that all values are numeric
for val in ts:
self.assertIsInstance(val, (int, float))
# Each time series text should be a string
for i, ts_text in enumerate(sample["time_series_text"]):
self.assertIsInstance(ts_text, str, f"Time series text {i} is not a string")
self.assertGreater(len(ts_text), 0, f"Time series text {i} is empty")
self.logger.success(f"Time series test passed: {len(sample['time_series'])} series")
def test_time_series_text_includes_ecg_info(self):
"""Test that each time_series_text includes ECG lead information."""
self.logger.info("Testing time series text format...")
sample = self.train_dataset[0]
for i, text in enumerate(sample['time_series_text']):
self.logger.debug(f"Testing text {i}: {text[:100]}...")
# Should mention ECG lead
self.assertIn('ECG Lead', text)
# Should mention mean and std
self.assertIn('mean', text)
self.assertIn('std', text)
# Should be in natural language format (optimized version uses simpler format)
# The optimized version uses: "This is ECG Lead X, it has mean Y and std Z:"
self.logger.success("Time series text format tests passed")
def test_labels_static_method(self):
"""Test that the static get_labels method returns expected labels."""
self.logger.info("Testing static get_labels method...")
labels = self.ECGQACoTQADataset.get_labels()
# Test that the method works and returns reasonable results
self.assertIsInstance(labels, list)
self.assertGreater(len(labels), 0, "Labels list should not be empty")
# Test that common labels are present
common_labels = ["yes", "no", "none"]
for label in common_labels:
self.assertIn(label, labels, f"Common label '{label}' should be present")
# Test that labels are strings
for label in labels:
self.assertIsInstance(label, str, f"All labels should be strings, got {type(label)}")
self.assertGreater(len(label.strip()), 0, "Labels should not be empty strings")
self.logger.success(f"Static labels test passed: {len(labels)} labels found")
def test_prompts_content(self):
"""Test that prompts contain expected content."""
self.logger.info("Testing prompt content...")
sample = self.train_dataset[0]
# Pre-prompt should mention cardiologist and ECG
pre_prompt = sample["pre_prompt"]
self.assertIn("cardiologist", pre_prompt.lower())
self.assertIn("ecg", pre_prompt.lower())
# Should mention the specific question
question = sample["question"]
self.assertIn(question, pre_prompt)
# Post-prompt should mention answer format
post_prompt = sample["post_prompt"]
self.assertIn("Answer:", post_prompt)
self.logger.success("Prompt content tests passed")
def test_cot_specific_fields(self):
"""Test that CoT-specific fields are preserved in the formatted sample."""
self.logger.info("Testing CoT-specific fields preservation...")
sample = self.train_dataset[0]
# Check fields that actually exist in the implementation
cot_fields = ["rationale", "template_id", "question_type"]
for field in cot_fields:
if field in sample:
self.logger.debug(f"Field {field}: {sample[field]}")
self.logger.success("CoT fields preservation tests passed")
def test_exclude_comparison_functionality(self):
"""Test that exclude_comparison properly filters out comparison questions."""
self.logger.info("Testing exclude_comparison functionality...")
# Check that the non-comparison dataset has no comparison questions
for sample in self.train_no_comparison:
question_type = sample.get("question_type")
if question_type is None:
raise ValueError(f"Sample missing question_type: {sample}")
self.assertFalse(question_type.startswith("comparison"),
f"Found comparison question in non-comparison dataset: {question_type}")
# Check that the regular dataset might have comparison questions
has_comparison = False
for sample in self.train_dataset:
question_type = sample.get("question_type")
if question_type and question_type.startswith("comparison"):
has_comparison = True
break
self.logger.success(f"Exclude comparison test passed. Regular dataset has comparison questions: {has_comparison}")
def test_100hz_data_consistency(self):
"""Test that all time series data is consistently 100Hz."""
self.logger.info("Testing 100Hz data consistency...")
for dataset_name, dataset in [("train", self.train_dataset), ("val", self.val_dataset), ("test", self.test_dataset)]:
for i, sample in enumerate(dataset):
if i >= 3: # Only test first 3 samples per dataset
break
time_series = sample.get("time_series")
if time_series is None:
continue
for j, ts in enumerate(time_series):
# Each time series should be exactly 1000 samples (10 seconds at 100Hz)
expected_length = 1000
actual_length = len(ts)
if actual_length != expected_length:
self.logger.warning(f"Dataset {dataset_name}, sample {i}, time series {j}: "
f"Expected {expected_length} samples, got {actual_length}")
# All values should be numeric
for val in ts:
self.assertIsInstance(val, (int, float),
f"Non-numeric value in time series: {val}")
self.logger.success("100Hz data consistency tests passed")
def test_error_handling_missing_fields(self):
"""Test that the dataset properly raises errors for missing required fields."""
self.logger.info("Testing error handling for missing fields...")
# Create a sample with missing fields to test error handling
from opentslm.time_series_datasets.ecg_qa import ECGQACoTQADataset
# This should work normally
try:
normal_dataset = ECGQACoTQADataset(split="train", EOS_TOKEN="", max_samples=1)
self.logger.success("Normal dataset creation works")
except Exception as e:
self.logger.error(f"Unexpected error creating normal dataset: {e}")
raise
# Test that we get proper errors when accessing samples with missing data
# (This would happen if the underlying data is corrupted)
self.logger.info("Error handling tests completed")
def test_example_data_QA(self):
"""Print example data for ECGQACoTQADataset, showing all time series and text."""
sample = self.train_dataset[0]
self.logger.info("="*80)
self.logger.info("EXAMPLE ECGQACoTQADataset SAMPLE")
self.logger.info("="*80)
self.logger.info(f"Question: '{sample['question']}'")
self.logger.info(f"Question type: '{sample['question_type']}'")
pre_prompt_preview = sample['pre_prompt'][:200]
self.logger.info(f"Pre-prompt: '{pre_prompt_preview}'")
self.logger.info(f"Post-prompt: '{sample['post_prompt']}'")
answer_preview = sample['answer']
self.logger.info(f"Answer (rationale): '{answer_preview}'")
self.logger.info(f"Number of time series: {len(sample['time_series'])}")
for i, (ts, ts_text) in enumerate(zip(sample['time_series'], sample['time_series_text'])):
self.logger.info(f"Time series {i} text: '{ts_text}'")
self.logger.info(f"Time series {i} length: {len(ts)}")
self.logger.info(f"First 10 values: {ts[:10]}")
self.logger.info(f"Last 10 values: {ts[-10:]}")
self.logger.info("="*80)
if __name__ == "__main__":
print("Running ECG-QA CoT loader and dataset tests, this might take a while...")
unittest.main()
|