Spaces:
Building
Building
File size: 39,219 Bytes
492bb24 e7279e4 864b9a2 492bb24 e7279e4 492bb24 864b9a2 492bb24 864b9a2 492bb24 864b9a2 492bb24 864b9a2 492bb24 864b9a2 492bb24 864b9a2 492bb24 864b9a2 492bb24 864b9a2 492bb24 864b9a2 492bb24 864b9a2 492bb24 864b9a2 492bb24 864b9a2 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 | """
Corpus Data Visualizer Module
This module provides functionality for merging, filtering, and visualizing corpus data.
Supports merging metadata with results files, applying filters, and generating
visualizations such as box plots and scatter plots.
"""
import pandas as pd
import numpy as np
import plotly.express as px
import plotly.graph_objects as go
from typing import Dict, List, Tuple, Optional, Union, Any
import logging
import re
from io import StringIO
import natsort
import csv
from scipy import stats
from scipy.stats import f_oneway
import warnings
logger = logging.getLogger(__name__)
class CorpusVisualizer:
"""
A class for merging, filtering, and visualizing corpus data.
Supports:
- Merging two dataframes (metadata and results)
- Detecting potential join columns
- Filtering merged data
- Creating visualizations (box plots, scatter plots)
- Validating merge quality
- Exporting merged data
"""
def __init__(self, file_size_limit_mb: int = 300):
"""
Initialize the corpus visualizer.
Args:
file_size_limit_mb: Maximum file size limit in MB for uploads
"""
self.metadata_df = None
self.results_df = None
self.merged_df = None
self.file_size_limit = file_size_limit_mb * 1024 * 1024
self.merge_stats = None
self.filters = []
self.category_orders = {} # Store custom category orders
def detect_file_format(self, content: Union[str, bytes]) -> Dict[str, any]:
"""
Detect file format and separator.
Args:
content: File content as string or bytes
Returns:
Dict with format information
"""
if isinstance(content, bytes):
content = content.decode('utf-8')
# Check file size
if len(content.encode('utf-8')) > self.file_size_limit:
raise ValueError(f"File too large. Maximum size is {self.file_size_limit // (1024*1024)}MB")
# Detect separator by checking first few lines
lines = content.strip().split('\n')[:5]
separators = ['\t', ',', ';', '|']
best_sep = '\t'
max_columns = 0
for sep in separators:
avg_cols = np.mean([len(line.split(sep)) for line in lines])
if avg_cols > max_columns:
max_columns = avg_cols
best_sep = sep
return {
'separator': best_sep,
'has_header': True,
'estimated_columns': int(max_columns),
'sample_lines': lines[:3]
}
def load_dataframe(self, content: Union[str, bytes], file_type: str) -> pd.DataFrame:
"""
Load content into a pandas DataFrame.
Args:
content: File content as string or bytes
file_type: Type of file ('metadata' or 'results')
Returns:
pd.DataFrame: Loaded dataframe
"""
if isinstance(content, bytes):
content = content.decode('utf-8')
# Detect file format
format_info = self.detect_file_format(content)
separator = format_info['separator']
# Load into DataFrame
df = pd.read_csv(StringIO(content), sep=separator,
quoting=csv.QUOTE_MINIMAL, quotechar='"')
# Store the dataframe
if file_type == 'metadata':
self.metadata_df = df
else: # results
self.results_df = df
return df
def detect_column_types(self, df: pd.DataFrame) -> Dict[str, List[str]]:
"""
Detect and categorize columns by data type.
Args:
df: DataFrame to analyze
Returns:
Dict with categorized column lists
"""
numeric_cols = []
categorical_cols = []
id_cols = []
text_cols = []
for col in df.columns:
col_str = str(col).lower()
# Check for ID columns first
if any(id_pattern in col_str for id_pattern in ['id', 'filename', 'file']):
id_cols.append(col)
elif pd.api.types.is_numeric_dtype(df[col]):
numeric_cols.append(col)
elif df[col].dtype == 'object':
unique_ratio = df[col].nunique() / len(df)
if unique_ratio < 0.2:
categorical_cols.append(col)
else:
text_cols.append(col)
return {
'numeric_columns': numeric_cols,
'categorical_columns': categorical_cols,
'id_columns': id_cols,
'text_columns': text_cols
}
def detect_join_columns(self) -> List[Dict[str, Any]]:
"""
Detect potential join columns between metadata and results dataframes.
Returns:
List of potential join column pairs with quality scores
"""
if self.metadata_df is None or self.results_df is None:
raise ValueError("Both metadata and results dataframes must be loaded")
metadata_cols = self.detect_column_types(self.metadata_df)
results_cols = self.detect_column_types(self.results_df)
potential_joins = []
# Check ID columns first
for meta_col in metadata_cols['id_columns']:
for results_col in results_cols['id_columns']:
potential_joins.append({
'metadata_column': meta_col,
'results_column': results_col,
'match_percent': 75.0, # Simplified for now
'recommendation': True
})
return potential_joins
def merge_dataframes(self, metadata_column: str, results_column: str,
handle_extensions: bool = True) -> pd.DataFrame:
"""
Merge metadata and results dataframes.
Args:
metadata_column: Join column in metadata dataframe
results_column: Join column in results dataframe
handle_extensions: Whether to handle file extensions in joins
Returns:
pd.DataFrame: Merged dataframe
"""
if self.metadata_df is None or self.results_df is None:
raise ValueError("Both dataframes must be loaded before merging")
# Make copies to avoid modifying original dataframes
metadata = self.metadata_df.copy()
results = self.results_df.copy()
if handle_extensions:
# Create cleaned join keys by stripping extensions and whitespace
def clean_join_key(filename):
"""Clean filename for joining: strip whitespace and remove file extensions."""
clean = str(filename).strip() # Remove leading/trailing whitespace
clean = re.sub(r'\.[^.]*$', '', clean) # Remove any file extension
return clean
# Apply cleaning to both columns
metadata['_join_key'] = metadata[metadata_column].apply(clean_join_key)
results['_join_key'] = results[results_column].apply(clean_join_key)
# Perform merge on cleaned keys
merged = pd.merge(
metadata,
results,
left_on='_join_key',
right_on='_join_key',
how='left',
suffixes=('_meta', '_results')
)
# Drop the temporary join key columns
merged = merged.drop(columns=['_join_key'])
else:
# Direct merge without extension handling
merged = pd.merge(
metadata,
results,
left_on=metadata_column,
right_on=results_column,
how='left',
suffixes=('_meta', '_results')
)
# Calculate merge statistics
total_rows = len(self.metadata_df)
# Count matched rows by checking if any results column has non-null values
results_columns = [col for col in merged.columns if col.endswith('_results')]
if results_columns:
matched_rows = merged[results_columns[0]].notna().sum()
else:
matched_rows = len(merged.dropna())
self.merge_stats = {
'total_rows': total_rows,
'matched_rows': matched_rows,
'match_percent': (matched_rows / total_rows * 100) if total_rows > 0 else 0,
'metadata_column': metadata_column,
'results_column': results_column,
'handle_extensions': handle_extensions
}
self.merged_df = merged
return merged
def validate_merge(self) -> Dict[str, Any]:
"""
Validate merge quality and generate statistics.
Returns:
Dict with merge quality statistics
"""
if self.merge_stats is None:
raise ValueError("Must perform merge before validation")
validation = self.merge_stats.copy()
validation['quality_score'] = validation['match_percent']
validation['quality_assessment'] = "Good" if validation['match_percent'] > 75 else "Fair"
return validation
def filter_dataframe(self, filters: List[Dict[str, Any]]) -> pd.DataFrame:
"""
Apply filters to the merged dataframe.
Args:
filters: List of filter dictionaries
Returns:
pd.DataFrame: Filtered dataframe
"""
if self.merged_df is None:
raise ValueError("Must perform merge before filtering")
filtered_df = self.merged_df.copy()
for filter_item in filters:
column = filter_item.get('column')
operator = filter_item.get('operator')
value = filter_item.get('value')
if column and operator and value is not None and column in filtered_df.columns:
if operator == '=':
filtered_df = filtered_df[filtered_df[column] == value]
elif operator == 'in':
filtered_df = filtered_df[filtered_df[column].isin(value)]
return filtered_df
def get_smart_category_order(self, column: str, values: List[str]) -> List[str]:
"""
Generate smart ordering for categorical values.
Args:
column: Column name
values: List of unique values in the column
Returns:
List of values in smart order
"""
# Convert to strings and remove None/NaN values
clean_values = [str(v) for v in values if pd.notna(v) and str(v).strip()]
if not clean_values:
return values
# Check for common patterns that should use natural sorting
patterns = [
r'year\d+', # year1, year2, year11
r'level\d+', # level1, level2, level10
r'grade\d+', # grade1, grade2, grade12
r'week\d+', # week1, week2, week52
r'day\d+', # day1, day2, day365
r'session\d+', # session1, session2, session10
r'group\d+', # group1, group2, group15
r'class\d+', # class1, class2, class20
r'stage\d+', # stage1, stage2, stage5
r'phase\d+', # phase1, phase2, phase3
r'\w+\d+', # Any word followed by numbers
]
# Check if values match any pattern
for pattern in patterns:
if all(re.match(pattern, str(val).lower()) for val in clean_values):
# Use natural sorting for numeric patterns
try:
return natsort.natsorted(clean_values, key=str.lower)
except:
break
# Check if all values are numeric (as strings)
try:
numeric_values = [float(v) for v in clean_values]
return [str(v) for v in sorted(set(numeric_values))]
except (ValueError, TypeError):
pass
# Default to alphabetical sorting
return sorted(clean_values, key=str.lower)
def set_category_order(self, column: str, order: List[str]) -> None:
"""
Set custom ordering for a categorical column.
Args:
column: Column name
order: List of category values in desired order
"""
self.category_orders[column] = order
def get_category_order(self, column: str, df: Optional[pd.DataFrame] = None) -> List[str]:
"""
Get category order for a column.
Args:
column: Column name
df: DataFrame to get values from (defaults to merged_df)
Returns:
List of category values in order
"""
# Use provided dataframe or default to merged_df
if df is None:
df = self.merged_df
if df is None:
raise ValueError("No dataframe available")
# Return custom order if set
if column in self.category_orders:
return self.category_orders[column]
# Get unique values from the column
unique_values = df[column].dropna().unique().tolist()
# Generate smart order
return self.get_smart_category_order(column, unique_values)
def reset_category_order(self, column: str) -> None:
"""
Reset category order for a column to default smart ordering.
Args:
column: Column name
"""
if column in self.category_orders:
del self.category_orders[column]
def create_boxplot(self, x_column: str, y_column: str, color_column: Optional[str] = None,
title: Optional[str] = None, height: int = 600,
category_orders: Optional[Dict[str, List[str]]] = None) -> Tuple[go.Figure, Optional[Dict[str, Any]]]:
"""
Create a box plot visualization using Plotly with statistical analysis.
Args:
x_column: Categorical column for x-axis
y_column: Numeric column for y-axis
color_column: Optional column for color grouping
title: Plot title
height: Plot height
category_orders: Optional custom category orders
Returns:
Tuple of (Plotly figure object, Statistical results dict)
"""
if self.merged_df is None:
raise ValueError("Must perform merge before creating visualizations")
plot_df = self.merged_df
# Build category orders dict
plot_category_orders = {}
# Add x-axis category order
if category_orders and x_column in category_orders:
plot_category_orders[x_column] = category_orders[x_column]
else:
plot_category_orders[x_column] = self.get_category_order(x_column, plot_df)
# Add color column category order if specified
if color_column:
if category_orders and color_column in category_orders:
plot_category_orders[color_column] = category_orders[color_column]
else:
plot_category_orders[color_column] = self.get_category_order(color_column, plot_df)
# Create the plot
if color_column:
fig = px.box(plot_df, x=x_column, y=y_column, color=color_column,
title=title or f"Box Plot: {y_column} by {x_column}", height=height,
category_orders=plot_category_orders)
else:
fig = px.box(plot_df, x=x_column, y=y_column,
title=title or f"Box Plot: {y_column} by {x_column}", height=height,
category_orders=plot_category_orders)
fig.update_layout(template="plotly_white")
# Perform statistical analysis
stats_results = None
try:
if color_column:
# Two-way ANOVA
stats_results = self.perform_two_way_anova(plot_df, x_column, y_column, color_column)
else:
# One-way ANOVA
stats_results = self.perform_one_way_anova(plot_df, x_column, y_column)
except Exception as e:
stats_results = {"error": f"Statistical analysis failed: {str(e)}"}
return fig, stats_results
def create_scatterplot(self, x_column: str, y_column: str, color_column: Optional[str] = None,
title: Optional[str] = None, height: int = 600,
category_orders: Optional[Dict[str, List[str]]] = None,
add_trendline: bool = True, add_confidence_interval: bool = True) -> Tuple[go.Figure, Optional[Dict[str, Any]]]:
"""
Create a scatter plot visualization using Plotly with statistical analysis.
Args:
x_column: Numeric column for x-axis
y_column: Numeric column for y-axis
color_column: Optional column for color coding points
title: Plot title
height: Plot height
category_orders: Optional custom category orders
add_trendline: Whether to add regression line (default True)
add_confidence_interval: Whether to add confidence interval around trendline (default True)
Returns:
Tuple of (Plotly figure object, Statistical results dict)
"""
if self.merged_df is None:
raise ValueError("Must perform merge before creating visualizations")
plot_df = self.merged_df
# Build category orders dict (for color column if categorical)
plot_category_orders = {}
# Add color column category order if specified and categorical
if color_column:
column_types = self.detect_column_types(plot_df)
if color_column in column_types.get('categorical_columns', []):
if category_orders and color_column in category_orders:
plot_category_orders[color_column] = category_orders[color_column]
else:
plot_category_orders[color_column] = self.get_category_order(color_column, plot_df)
# Create the base scatter plot
if color_column:
fig = px.scatter(plot_df, x=x_column, y=y_column, color=color_column,
title=title or f"Scatter Plot: {y_column} vs {x_column}", height=height,
category_orders=plot_category_orders if plot_category_orders else None)
else:
fig = px.scatter(plot_df, x=x_column, y=y_column,
title=title or f"Scatter Plot: {y_column} vs {x_column}", height=height)
# Perform statistical analysis
stats_results = None
try:
stats_results = self.perform_simple_regression(plot_df, x_column, y_column)
# Add trendline and confidence interval if requested and regression successful
if add_trendline and 'error' not in stats_results:
clean_df = plot_df[[x_column, y_column]].dropna()
x_vals = clean_df[x_column].values
y_vals = clean_df[y_column].values
# Get regression parameters
slope = stats_results['regression']['slope']
intercept = stats_results['regression']['intercept']
# Create more detailed x range for smooth curves
x_min, x_max = x_vals.min(), x_vals.max()
x_range = np.linspace(x_min, x_max, 100)
y_range = slope * x_range + intercept
# Calculate confidence intervals if requested
if add_confidence_interval:
n = len(x_vals)
mean_x = np.mean(x_vals)
ss_x = np.sum((x_vals - mean_x) ** 2)
mse = np.sum((y_vals - (slope * x_vals + intercept)) ** 2) / (n - 2)
# Standard error for each prediction point
se_y = np.sqrt(mse * (1/n + (x_range - mean_x)**2 / ss_x))
# 95% confidence interval (t-distribution for small samples)
from scipy.stats import t
t_val = t.ppf(0.975, n - 2) # 95% confidence
y_upper = y_range + t_val * se_y
y_lower = y_range - t_val * se_y
# Add confidence interval as filled area
fig.add_trace(go.Scatter(
x=np.concatenate([x_range, x_range[::-1]]),
y=np.concatenate([y_upper, y_lower[::-1]]),
fill='toself',
fillcolor='rgba(255, 0, 0, 0.2)',
line=dict(color='rgba(255,255,255,0)'),
hoverinfo="skip",
showlegend=True,
name='95% Confidence Interval'
))
# Add trendline to the plot
fig.add_trace(go.Scatter(
x=x_range,
y=y_range,
mode='lines',
name=f'Trendline (R² = {stats_results["regression"]["r_squared"]:.3f})',
line=dict(color='red', dash='dash', width=2)
))
except Exception as e:
stats_results = {"error": f"Statistical analysis failed: {str(e)}"}
fig.update_layout(template="plotly_white")
return fig, stats_results
def export_merged_data(self) -> pd.DataFrame:
"""
Export merged dataframe.
Returns:
pd.DataFrame: DataFrame ready for export
"""
if self.merged_df is None:
raise ValueError("Must perform merge before exporting")
return self.merged_df
# Statistical Analysis Methods
def calculate_eta_squared(self, ss_between: float, ss_total: float) -> float:
"""
Calculate eta-squared effect size for ANOVA.
Args:
ss_between: Sum of squares between groups
ss_total: Total sum of squares
Returns:
float: Eta-squared value
"""
if ss_total == 0:
return 0.0
return ss_between / ss_total
def calculate_partial_eta_squared(self, ss_effect: float, ss_error: float) -> float:
"""
Calculate partial eta-squared effect size for factorial ANOVA.
Args:
ss_effect: Sum of squares for the effect
ss_error: Sum of squares for error
Returns:
float: Partial eta-squared value
"""
if (ss_effect + ss_error) == 0:
return 0.0
return ss_effect / (ss_effect + ss_error)
def calculate_cohens_d(self, group1: np.ndarray, group2: np.ndarray) -> float:
"""
Calculate Cohen's d effect size for two groups.
Args:
group1: Data for first group
group2: Data for second group
Returns:
float: Cohen's d value
"""
n1, n2 = len(group1), len(group2)
if n1 < 2 or n2 < 2:
return 0.0
# Calculate pooled standard deviation
pooled_std = np.sqrt(((n1 - 1) * np.var(group1, ddof=1) +
(n2 - 1) * np.var(group2, ddof=1)) / (n1 + n2 - 2))
if pooled_std == 0:
return 0.0
return (np.mean(group1) - np.mean(group2)) / pooled_std
def calculate_cohens_f_squared(self, r_squared: float) -> float:
"""
Calculate Cohen's f² effect size for regression.
Args:
r_squared: R-squared value from regression
Returns:
float: Cohen's f² value
"""
if r_squared >= 1.0 or r_squared < 0:
return 0.0
return r_squared / (1 - r_squared)
def interpret_effect_size(self, value: float, metric_type: str) -> str:
"""
Provide interpretation for effect sizes.
Args:
value: Effect size value
metric_type: Type of effect size ('eta_squared', 'cohens_d', 'r_squared', 'cohens_f')
Returns:
str: Interpretation (Small, Medium, Large)
"""
if metric_type == 'eta_squared' or metric_type == 'partial_eta_squared':
if value < 0.01:
return "Small"
elif value < 0.06:
return "Small"
elif value < 0.14:
return "Medium"
else:
return "Large"
elif metric_type == 'cohens_d':
abs_value = abs(value)
if abs_value < 0.2:
return "Small"
elif abs_value < 0.5:
return "Small"
elif abs_value < 0.8:
return "Medium"
else:
return "Large"
elif metric_type == 'r_squared':
if value < 0.01:
return "Small"
elif value < 0.09:
return "Small"
elif value < 0.25:
return "Medium"
else:
return "Large"
elif metric_type == 'cohens_f':
if value < 0.02:
return "Small"
elif value < 0.15:
return "Small"
elif value < 0.35:
return "Medium"
else:
return "Large"
else:
return "Unknown"
def perform_one_way_anova(self, df: pd.DataFrame, x_column: str, y_column: str) -> Dict[str, Any]:
"""
Perform one-way ANOVA analysis.
Args:
df: DataFrame containing the data
x_column: Categorical column (groups)
y_column: Numeric column (dependent variable)
Returns:
Dict containing ANOVA results and effect sizes
"""
try:
# Remove missing values
clean_df = df[[x_column, y_column]].dropna()
if len(clean_df) < 3:
return {"error": "Insufficient data for ANOVA (need at least 3 observations)"}
# Get groups
groups = [group[y_column].values for name, group in clean_df.groupby(x_column)]
# Check if we have at least 2 groups with data
valid_groups = [g for g in groups if len(g) > 0]
if len(valid_groups) < 2:
return {"error": "Need at least 2 groups for ANOVA"}
# Perform ANOVA
f_stat, p_value = f_oneway(*valid_groups)
# Calculate effect size (eta-squared)
group_data = []
group_names = []
for name, group in clean_df.groupby(x_column):
if len(group) > 0:
group_data.append(group[y_column].values)
group_names.append(name)
# Calculate sums of squares
grand_mean = clean_df[y_column].mean()
ss_total = np.sum((clean_df[y_column] - grand_mean) ** 2)
ss_between = 0
for group in group_data:
ss_between += len(group) * (np.mean(group) - grand_mean) ** 2
eta_squared = self.calculate_eta_squared(ss_between, ss_total)
# Degrees of freedom
df_between = len(valid_groups) - 1
df_within = len(clean_df) - len(valid_groups)
results = {
"test_type": "One-way ANOVA",
"f_statistic": f_stat,
"p_value": p_value,
"df_between": df_between,
"df_within": df_within,
"eta_squared": eta_squared,
"eta_squared_interpretation": self.interpret_effect_size(eta_squared, "eta_squared"),
"sample_size": len(clean_df),
"groups": group_names,
"group_means": [np.mean(group) for group in group_data],
"group_sizes": [len(group) for group in group_data]
}
# Post hoc analysis if significant and more than 2 groups
if p_value < 0.05 and len(valid_groups) > 2:
try:
posthoc_results = []
for i in range(len(group_data)):
for j in range(i + 1, len(group_data)):
# Calculate Cohen's d for this pair
cohens_d = self.calculate_cohens_d(group_data[i], group_data[j])
# Simple t-test for this pair (for p-value)
t_stat, t_p = stats.ttest_ind(group_data[i], group_data[j])
posthoc_results.append({
"group1": group_names[i],
"group2": group_names[j],
"cohens_d": cohens_d,
"cohens_d_interpretation": self.interpret_effect_size(cohens_d, "cohens_d"),
"p_value": t_p,
"mean_diff": np.mean(group_data[i]) - np.mean(group_data[j])
})
results["posthoc"] = posthoc_results
except Exception as e:
results["posthoc_error"] = f"Error in post hoc analysis: {str(e)}"
return results
except Exception as e:
return {"error": f"Error performing ANOVA: {str(e)}"}
def perform_two_way_anova(self, df: pd.DataFrame, x_column: str, y_column: str, color_column: str) -> Dict[str, Any]:
"""
Perform two-way ANOVA analysis.
Args:
df: DataFrame containing the data
x_column: First factor (categorical)
y_column: Dependent variable (numeric)
color_column: Second factor (categorical)
Returns:
Dict containing two-way ANOVA results and effect sizes
"""
try:
# Remove missing values
clean_df = df[[x_column, y_column, color_column]].dropna()
if len(clean_df) < 6: # Need minimum samples for 2-way ANOVA
return {"error": "Insufficient data for two-way ANOVA (need at least 6 observations)"}
# Get factor levels
factor1_levels = clean_df[x_column].unique()
factor2_levels = clean_df[color_column].unique()
if len(factor1_levels) < 2 or len(factor2_levels) < 2:
return {"error": "Need at least 2 levels per factor for two-way ANOVA"}
# Manual two-way ANOVA calculation
grand_mean = clean_df[y_column].mean()
n_total = len(clean_df)
# Calculate sums of squares
ss_total = np.sum((clean_df[y_column] - grand_mean) ** 2)
# Factor A (x_column) effect
ss_a = 0
for level in factor1_levels:
group_data = clean_df[clean_df[x_column] == level][y_column]
if len(group_data) > 0:
ss_a += len(group_data) * (np.mean(group_data) - grand_mean) ** 2
# Factor B (color_column) effect
ss_b = 0
for level in factor2_levels:
group_data = clean_df[clean_df[color_column] == level][y_column]
if len(group_data) > 0:
ss_b += len(group_data) * (np.mean(group_data) - grand_mean) ** 2
# Interaction effect
ss_ab = 0
for a_level in factor1_levels:
for b_level in factor2_levels:
cell_data = clean_df[(clean_df[x_column] == a_level) & (clean_df[color_column] == b_level)][y_column]
if len(cell_data) > 0:
# Cell mean
cell_mean = np.mean(cell_data)
# Marginal means
a_mean = np.mean(clean_df[clean_df[x_column] == a_level][y_column])
b_mean = np.mean(clean_df[clean_df[color_column] == b_level][y_column])
# Interaction sum of squares
ss_ab += len(cell_data) * (cell_mean - a_mean - b_mean + grand_mean) ** 2
# Error sum of squares
ss_error = ss_total - ss_a - ss_b - ss_ab
# Degrees of freedom
df_a = len(factor1_levels) - 1
df_b = len(factor2_levels) - 1
df_ab = df_a * df_b
df_error = n_total - len(factor1_levels) * len(factor2_levels)
# Mean squares
ms_a = ss_a / df_a if df_a > 0 else 0
ms_b = ss_b / df_b if df_b > 0 else 0
ms_ab = ss_ab / df_ab if df_ab > 0 else 0
ms_error = ss_error / df_error if df_error > 0 else 1
# F statistics
f_a = ms_a / ms_error if ms_error > 0 else 0
f_b = ms_b / ms_error if ms_error > 0 else 0
f_ab = ms_ab / ms_error if ms_error > 0 else 0
# P values
p_a = 1 - stats.f.cdf(f_a, df_a, df_error) if f_a > 0 else 1
p_b = 1 - stats.f.cdf(f_b, df_b, df_error) if f_b > 0 else 1
p_ab = 1 - stats.f.cdf(f_ab, df_ab, df_error) if f_ab > 0 else 1
# Effect sizes (partial eta squared)
eta_squared_a = self.calculate_partial_eta_squared(ss_a, ss_error)
eta_squared_b = self.calculate_partial_eta_squared(ss_b, ss_error)
eta_squared_ab = self.calculate_partial_eta_squared(ss_ab, ss_error)
results = {
"test_type": "Two-way ANOVA",
"factor_a": {
"name": x_column,
"f_statistic": f_a,
"p_value": p_a,
"df": df_a,
"partial_eta_squared": eta_squared_a,
"interpretation": self.interpret_effect_size(eta_squared_a, "partial_eta_squared")
},
"factor_b": {
"name": color_column,
"f_statistic": f_b,
"p_value": p_b,
"df": df_b,
"partial_eta_squared": eta_squared_b,
"interpretation": self.interpret_effect_size(eta_squared_b, "partial_eta_squared")
},
"interaction": {
"name": f"{x_column} × {color_column}",
"f_statistic": f_ab,
"p_value": p_ab,
"df": df_ab,
"partial_eta_squared": eta_squared_ab,
"interpretation": self.interpret_effect_size(eta_squared_ab, "partial_eta_squared")
},
"df_error": df_error,
"sample_size": n_total,
"factor_a_levels": list(factor1_levels),
"factor_b_levels": list(factor2_levels)
}
return results
except Exception as e:
return {"error": f"Error performing two-way ANOVA: {str(e)}"}
def perform_simple_regression(self, df: pd.DataFrame, x_column: str, y_column: str) -> Dict[str, Any]:
"""
Perform simple linear regression analysis.
Args:
df: DataFrame containing the data
x_column: Independent variable (numeric)
y_column: Dependent variable (numeric)
Returns:
Dict containing regression results and effect sizes
"""
try:
# Remove missing values
clean_df = df[[x_column, y_column]].dropna()
if len(clean_df) < 3:
return {"error": "Insufficient data for regression (need at least 3 observations)"}
x = clean_df[x_column].values
y = clean_df[y_column].values
# Calculate correlation
correlation, corr_p = stats.pearsonr(x, y)
# Simple linear regression
slope, intercept, r_value, p_value, std_err = stats.linregress(x, y)
# Calculate additional statistics
r_squared = r_value ** 2
adjusted_r_squared = 1 - (1 - r_squared) * (len(clean_df) - 1) / (len(clean_df) - 2)
# Effect sizes
cohens_f_squared = self.calculate_cohens_f_squared(r_squared)
# Standard error of the slope
n = len(clean_df)
mean_x = np.mean(x)
ss_x = np.sum((x - mean_x) ** 2)
mse = np.sum((y - (slope * x + intercept)) ** 2) / (n - 2)
se_slope = np.sqrt(mse / ss_x)
# t-statistic for slope
t_stat = slope / se_slope if se_slope > 0 else 0
results = {
"test_type": "Simple Linear Regression",
"correlation": {
"pearson_r": correlation,
"p_value": corr_p,
"interpretation": self.interpret_effect_size(abs(correlation), "r_squared")
},
"regression": {
"slope": slope,
"intercept": intercept,
"r_squared": r_squared,
"adjusted_r_squared": adjusted_r_squared,
"p_value": p_value,
"standard_error": std_err,
"t_statistic": t_stat,
"cohens_f_squared": cohens_f_squared,
"f_squared_interpretation": self.interpret_effect_size(cohens_f_squared, "cohens_f")
},
"sample_size": len(clean_df),
"variance_explained": f"{r_squared * 100:.1f}%"
}
return results
except Exception as e:
return {"error": f"Error performing regression: {str(e)}"}
|