File size: 2,675 Bytes
8481f3c | 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 | import os
import pandas as pd
from tqdm import tqdm
def validate_csv(filepath):
"""Updated validation with more flexible timestamp checking"""
try:
df = pd.read_csv(filepath)
# Basic checks
assert not df.empty, "Empty DataFrame"
required_columns = {'timestamp', 'open', 'high', 'low', 'close', 'volume'}
assert required_columns.issubset(df.columns), f"Missing columns: {required_columns - set(df.columns)}"
# Data quality checks
duplicates = df.duplicated().sum()
assert duplicates == 0, f"{duplicates} duplicates found"
nulls = df.isnull().sum().sum()
assert nulls == 0, f"{nulls} null values found"
# Convert timestamps
df['timestamp'] = pd.to_datetime(df['timestamp'])
# Check OHLC relationships
assert (df['high'] >= df['low']).all(), "High < Low found"
assert (df['high'] >= df['open']).all(), "High < Open found"
assert (df['high'] >= df['close']).all(), "High < Close found"
assert (df['low'] <= df['open']).all(), "Low > Open found"
assert (df['low'] <= df['close']).all(), "Low > Close found"
# Check for major gaps in timeline (more flexible than strict order)
time_diff = df['timestamp'].diff().dt.total_seconds()
if (time_diff < 0).any():
print(f"Warning: {filepath} has {sum(time_diff < 0)} timestamp decreases")
return True, "Validation passed with warnings"
except Exception as e:
return False, str(e)
def validate_dataset(data_dir="data"):
"""Validate all CSV files in a directory."""
results = {}
for root, _, files in os.walk(data_dir):
for file in tqdm(files, desc="Validating files"):
if file.endswith('.csv'):
filepath = os.path.join(root, file)
is_valid, message = validate_csv(filepath)
results[file] = {
"valid": is_valid,
"message": message,
"path": filepath
}
# Print summary
print("\nValidation Summary:")
print("-" * 50)
for file, result in results.items():
status = "PASSED" if result["valid"] else "FAILED"
print(f"{file:<50} {status:<10} {result['message']}")
return results
if __name__ == "__main__":
# Example usage
validation_results = validate_dataset()
# Save validation report
pd.DataFrame.from_dict(validation_results, orient='index').to_csv("validation_report.csv")
print("\nValidation report saved to validation_report.csv") |