Spaces:
Sleeping
Sleeping
File size: 29,046 Bytes
c42749b 6337cd5 c42749b 6a3c971 c42749b 6a3c971 c50f214 6a3c971 c42749b c50f214 6337cd5 c42749b 6a3c971 c50f214 6337cd5 c42749b 6a3c971 6337cd5 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 c42749b c50f214 6337cd5 c42749b c50f214 6337cd5 c50f214 c42749b 6a3c971 c50f214 c42749b c50f214 6a3c971 c42749b c50f214 6a3c971 c42749b c50f214 c42749b c50f214 c42749b c50f214 c42749b c50f214 c42749b c50f214 6a3c971 c42749b c50f214 6a3c971 c50f214 c42749b 6a3c971 c42749b 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 c42749b c50f214 86805f4 c50f214 6a3c971 c50f214 6a3c971 c50f214 c42749b c50f214 6a3c971 c50f214 6a3c971 c50f214 c42749b c50f214 6a3c971 c42749b c50f214 c42749b c50f214 86805f4 c50f214 c42749b c50f214 c42749b c50f214 86805f4 c50f214 86805f4 c50f214 86805f4 c50f214 c42749b c50f214 86805f4 c50f214 86805f4 c50f214 c42749b c50f214 c42749b c50f214 7583e80 c50f214 6a3c971 c50f214 6a3c971 c50f214 c42749b 6a3c971 c50f214 7583e80 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 86805f4 c50f214 c42749b c50f214 c42749b c50f214 7583e80 c50f214 6a3c971 c50f214 6a3c971 c50f214 c42749b c50f214 6a3c971 c50f214 6a3c971 c50f214 86805f4 c50f214 6a3c971 c50f214 6a3c971 c50f214 6337cd5 c50f214 c42749b c50f214 c42749b c50f214 c42749b c50f214 c42749b c50f214 c42749b c50f214 c42749b c50f214 6a3c971 c50f214 c42749b c50f214 c42749b c50f214 6a3c971 c50f214 86805f4 c50f214 86805f4 c50f214 e879f17 c50f214 86805f4 c50f214 86805f4 c50f214 c42749b c50f214 86805f4 c50f214 c42749b c50f214 c42749b c50f214 c42749b c50f214 86805f4 c50f214 86805f4 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c42749b c50f214 c42749b c50f214 c42749b c50f214 c42749b c50f214 c42749b c50f214 c42749b c50f214 6a3c971 c50f214 6a3c971 c50f214 7583e80 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 6a3c971 c50f214 c42749b c50f214 c42749b c50f214 | 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 | import streamlit as st
import pandas as pd
import numpy as np
import plotly.express as px
import plotly.graph_objects as go
from typing import Dict, List, Any, Optional
import os
from dotenv import load_dotenv
from data_handler import *
from io import BytesIO
# Load environment variables
load_dotenv()
# Optional AI Integration
try:
import openai
OPENAI_AVAILABLE = True
except ImportError:
OPENAI_AVAILABLE = False
try:
import google.generativeai as genai
GEMINI_AVAILABLE = True
except ImportError:
GEMINI_AVAILABLE = False
class AIAssistant:
"""AI-powered analysis assistant"""
def __init__(self):
self.openai_key = os.getenv('OPENAI_API_KEY')
self.gemini_key = os.getenv('GOOGLE_API_KEY')
if self.gemini_key and GEMINI_AVAILABLE:
genai.configure(api_key=self.gemini_key)
self.gemini_model = genai.GenerativeModel('gemini-1.5-flash')
def get_available_models(self) -> List[str]:
"""Get list of available AI models"""
models = []
if self.openai_key and OPENAI_AVAILABLE:
models.append("OpenAI GPT")
if self.gemini_key and GEMINI_AVAILABLE:
models.append("Google Gemini")
return models
def analyze_insights(self, df: pd.DataFrame, insights: List[Dict], model: str = "Google Gemini") -> str:
"""Get AI analysis of insights"""
# Prepare data summary
summary = f"""
Dataset Summary:
- Shape: {df.shape}
- Columns: {list(df.columns)}
- Data types: {df.dtypes.value_counts().to_dict()}
Key Insights Found:
"""
for insight in insights:
summary += f"\n- {insight['insight']}"
prompt = f"""
As a senior data scientist, analyze this dataset and provide:
1. Business implications of the findings
2. Potential opportunities or risks
3. Recommendations for decision-making
4. Suggestions for further analysis
{summary}
Provide actionable insights in a professional format.
"""
try:
if model == "Google Gemini" and hasattr(self, 'gemini_model'):
response = self.gemini_model.generate_content(prompt)
return response.text
elif model == "OpenAI GPT" and self.openai_key:
client = openai.OpenAI(api_key=self.openai_key)
response = client.chat.completions.create(
model="gpt-3.5-turbo",
messages=[{"role": "user", "content": prompt}]
)
return response.choices[0].message.content
else:
return "AI analysis not available. Please configure API keys."
except Exception as e:
return f"AI Analysis Error: {str(e)}"
class DataAnalysisWorkflow:
"""Optimized data analysis workflow with caching and pagination"""
def __init__(self, df: pd.DataFrame):
self.df = df
self.stats = calculate_basic_stats(df)
self.column_types = get_column_types(df)
self.insights = []
self.page_size = 1000 # For pagination
def add_insight(self, insight: str, stage: int):
"""Add insight to analysis report"""
self.insights.append({
'stage': stage,
'insight': insight,
'timestamp': pd.Timestamp.now()
})
def get_paginated_data(self, page: int = 0) -> pd.DataFrame:
"""Get paginated data for display"""
start_idx = page * self.page_size
end_idx = start_idx + self.page_size
return self.df.iloc[start_idx:end_idx]
def stage_1_overview(self):
"""Stage 1: Data Overview with caching"""
st.subheader("π Data Overview")
# Data Quality Score
quality_metrics = calculate_data_quality_score(self.df)
col1, col2, col3, col4 = st.columns(4)
with col1:
st.metric("Rows", f"{self.stats['shape'][0]:,}")
with col2:
st.metric("Columns", f"{self.stats['shape'][1]:,}")
with col3:
st.metric("Quality Score", f"{quality_metrics['score']:.1f}/100")
with col4:
st.metric("Grade", quality_metrics['grade'])
if quality_metrics['issues']:
st.warning("Quality Issues Found:")
for issue in quality_metrics['issues']:
st.write(f"β’ {issue}")
# Memory Usage and Optimization
st.subheader("Memory Analysis")
memory_opt = calculate_memory_optimization(self.df)
col1, col2 = st.columns(2)
with col1:
st.metric("Current Memory", f"{memory_opt['current_memory_mb']:.1f} MB")
with col2:
if memory_opt['potential_savings_mb'] > 0:
st.metric("Potential Savings",
f"{memory_opt['potential_savings_mb']:.1f} MB",
f"{memory_opt['potential_savings_pct']:.1f}%")
if st.button("Show Optimization Details"):
st.dataframe(pd.DataFrame(memory_opt['suggestions']))
# Column Cardinality Analysis
st.subheader("Column Cardinality Analysis")
cardinality_df = calculate_column_cardinality(self.df)
# Filter options
col_types = cardinality_df['Type'].unique()
selected_types = st.multiselect("Filter by Column Type",
col_types,
default=col_types)
filtered_df = cardinality_df[cardinality_df['Type'].isin(selected_types)]
st.dataframe(filtered_df, use_container_width=True)
# Highlight important findings
id_cols = filtered_df[filtered_df['Type'] == 'Unique Identifier']['Column'].tolist()
if id_cols:
st.info(f"π Potential ID columns found: {', '.join(id_cols)}")
const_cols = filtered_df[filtered_df['Type'] == 'Constant']['Column'].tolist()
if const_cols:
st.warning(f"β οΈ Constant columns found: {', '.join(const_cols)}")
# Data types visualization
if self.stats['dtypes']:
st.subheader("Data Types Distribution")
fig = px.pie(values=list(self.stats['dtypes'].values()),
names=list(self.stats['dtypes'].keys()),
title="Data Types")
st.plotly_chart(fig, use_container_width=True)
# Sample data with pagination
st.subheader("Sample Data")
total_pages = (len(self.df) - 1) // self.page_size + 1
if total_pages > 1:
page = st.slider("Page", 0, total_pages - 1, 0)
sample_data = self.get_paginated_data(page)
st.write(f"Showing rows {page * self.page_size + 1} to {min((page + 1) * self.page_size, len(self.df))}")
else:
sample_data = self.df.head(10)
st.dataframe(sample_data, use_container_width=True)
# Missing values analysis
missing_df = calculate_missing_data(self.df)
if not missing_df.empty:
st.subheader("Missing Values Analysis")
st.dataframe(missing_df, use_container_width=True)
worst_column = missing_df.iloc[0]['Column']
worst_percentage = missing_df.iloc[0]['Missing %']
self.add_insight(f"Column '{worst_column}' has highest missing data: {worst_percentage:.1f}%", 1)
else:
st.success("β
No missing values found!")
self.add_insight("Dataset has no missing values - excellent data quality", 1)
# Add insights about data quality and cardinality
if quality_metrics['score'] < 80:
self.add_insight(f"Data quality needs improvement (Score: {quality_metrics['score']:.1f}/100)", 1)
if memory_opt['potential_savings_pct'] > 20:
self.add_insight(f"Potential memory optimization of {memory_opt['potential_savings_pct']:.1f}% identified", 1)
if id_cols:
self.add_insight(f"Found {len(id_cols)} potential ID columns", 1)
def stage_2_exploration(self):
"""Stage 2: Exploratory Data Analysis with caching"""
st.subheader("π Exploratory Data Analysis")
numeric_cols = self.column_types['numeric']
categorical_cols = self.column_types['categorical']
# Numeric analysis
if numeric_cols:
st.subheader("Numeric Variables")
selected_numeric = st.selectbox("Select numeric column:", numeric_cols)
col1, col2 = st.columns(2)
with col1:
fig = px.histogram(self.df, x=selected_numeric,
title=f"Distribution of {selected_numeric}")
st.plotly_chart(fig, use_container_width=True)
with col2:
fig = px.box(self.df, y=selected_numeric,
title=f"Box Plot of {selected_numeric}")
st.plotly_chart(fig, use_container_width=True)
# Statistical summary
st.subheader("Statistical Summary")
summary_stats = self.df[numeric_cols].describe()
st.dataframe(summary_stats, use_container_width=True)
# Correlation analysis
if len(numeric_cols) > 1:
st.subheader("Correlation Analysis")
corr_matrix = calculate_correlation_matrix(self.df)
if not corr_matrix.empty:
fig = px.imshow(corr_matrix, text_auto=True, aspect="auto",
title="Correlation Matrix")
st.plotly_chart(fig, use_container_width=True)
# Find highest correlation
corr_values = []
for i in range(len(corr_matrix.columns)):
for j in range(i+1, len(corr_matrix.columns)):
corr_values.append(abs(corr_matrix.iloc[i, j]))
if corr_values:
max_corr = max(corr_values)
self.add_insight(f"Maximum correlation coefficient: {max_corr:.3f}", 2)
# Categorical analysis
if categorical_cols:
st.subheader("Categorical Variables")
selected_categorical = st.selectbox("Select categorical column:", categorical_cols)
value_counts = get_value_counts(self.df, selected_categorical)
fig = px.bar(x=value_counts.index, y=value_counts.values,
title=f"Top 10 {selected_categorical} Values")
st.plotly_chart(fig, use_container_width=True)
total_categories = self.df[selected_categorical].nunique()
self.add_insight(f"Column '{selected_categorical}' has {total_categories} unique categories", 2)
def stage_3_cleaning(self):
"""Stage 3: Data Quality Assessment"""
st.subheader("π§Ή Data Quality Assessment")
cleaning_actions = []
cleaning_history = []
# Missing values handling
if self.stats['missing_values'] > 0:
st.subheader("Missing Values Treatment")
missing_df = calculate_missing_data(self.df)
st.dataframe(missing_df, use_container_width=True)
col1, col2 = st.columns(2)
with col1:
selected_col = st.selectbox("Select column to handle missing values:",
missing_df['Column'].tolist())
with col2:
fill_method = st.selectbox("Choose fill method:",
["Drop rows", "Mean", "Median", "Mode", "Custom value"])
if st.button("Apply Missing Value Treatment"):
try:
if fill_method == "Drop rows":
self.df = self.df.dropna(subset=[selected_col])
cleaning_history.append(f"Dropped rows with missing values in {selected_col}")
else:
if fill_method == "Mean":
fill_value = self.df[selected_col].mean()
elif fill_method == "Median":
fill_value = self.df[selected_col].median()
elif fill_method == "Mode":
fill_value = self.df[selected_col].mode()[0]
else: # Custom value
fill_value = st.number_input("Enter custom value:", value=0.0)
self.df[selected_col] = self.df[selected_col].fillna(fill_value)
cleaning_history.append(f"Filled missing values in {selected_col} with {fill_method}")
st.success("β
Missing values handled successfully!")
except Exception as e:
st.error(f"Error handling missing values: {str(e)}")
# Duplicates handling
if self.stats['duplicates'] > 0:
st.subheader("Duplicate Rows")
st.warning(f"Found {self.stats['duplicates']} duplicate rows")
if st.button("Remove Duplicate Rows"):
original_len = len(self.df)
self.df = self.df.drop_duplicates()
removed = original_len - len(self.df)
cleaning_history.append(f"Removed {removed} duplicate rows")
st.success(f"β
Removed {removed} duplicate rows")
else:
st.success("β
No duplicate rows found")
# Mixed type detection and handling
mixed_types = detect_mixed_types(self.df)
if mixed_types:
st.subheader("Mixed Data Types")
mixed_df = pd.DataFrame(mixed_types)
st.dataframe(mixed_df, use_container_width=True)
selected_col = st.selectbox("Select column to fix data type:",
[item['column'] for item in mixed_types])
fix_method = st.selectbox("Choose fix method:",
["Convert to numeric", "Convert to string"])
if st.button("Fix Data Type"):
try:
if fix_method == "Convert to numeric":
self.df[selected_col] = pd.to_numeric(self.df[selected_col], errors='coerce')
else:
self.df[selected_col] = self.df[selected_col].astype(str)
cleaning_history.append(f"Fixed data type for {selected_col} to {fix_method}")
st.success("β
Data type fixed successfully!")
except Exception as e:
st.error(f"Error fixing data type: {str(e)}")
# Outlier detection and handling
numeric_cols = self.column_types['numeric']
if numeric_cols:
st.subheader("Outlier Detection")
selected_col = st.selectbox("Select column for outlier detection:", numeric_cols)
outliers = calculate_outliers(self.df, selected_col)
outlier_count = len(outliers)
if outlier_count > 0:
st.warning(f"Found {outlier_count} potential outliers in '{selected_col}'")
st.dataframe(outliers[[selected_col]].head(100), use_container_width=True)
treatment_method = st.selectbox("Choose outlier treatment method:",
["None", "Remove", "Cap at percentiles"])
if treatment_method != "None" and st.button("Apply Outlier Treatment"):
try:
if treatment_method == "Remove":
self.df = self.df[~self.df.index.isin(outliers.index)]
cleaning_history.append(f"Removed {outlier_count} outliers from {selected_col}")
else: # Cap at percentiles
Q1 = self.df[selected_col].quantile(0.25)
Q3 = self.df[selected_col].quantile(0.75)
IQR = Q3 - Q1
lower_bound = Q1 - 1.5 * IQR
upper_bound = Q3 + 1.5 * IQR
self.df[selected_col] = self.df[selected_col].clip(lower_bound, upper_bound)
cleaning_history.append(f"Capped outliers in {selected_col} at percentiles")
st.success("β
Outliers handled successfully!")
except Exception as e:
st.error(f"Error handling outliers: {str(e)}")
else:
st.success(f"β
No outliers detected in '{selected_col}'")
# Cleaning History
if cleaning_history:
st.subheader("Cleaning Operations History")
for i, operation in enumerate(cleaning_history, 1):
st.write(f"{i}. {operation}")
self.add_insight(f"Performed {len(cleaning_history)} data cleaning operations", 3)
# Summary
if cleaning_actions:
st.subheader("Remaining Action Items")
for i, action in enumerate(cleaning_actions, 1):
st.write(f"{i}. {action}")
self.add_insight(f"Identified {len(cleaning_actions)} data quality issues", 3)
else:
st.success("β
Data quality is excellent!")
self.add_insight("No major data quality issues found", 3)
def stage_4_analysis(self):
"""Stage 4: Advanced Analysis"""
st.subheader("π¬ Advanced Analysis")
numeric_cols = self.column_types['numeric']
categorical_cols = self.column_types['categorical']
# Relationship analysis
if len(numeric_cols) >= 2:
st.subheader("Variable Relationships")
col1, col2 = st.columns(2)
with col1:
x_var = st.selectbox("X Variable:", numeric_cols)
with col2:
y_var = st.selectbox("Y Variable:",
[col for col in numeric_cols if col != x_var])
# Sample data for performance if dataset is large
sample_size = min(5000, len(self.df))
sample_df = self.df.sample(n=sample_size) if len(self.df) > sample_size else self.df
fig = px.scatter(sample_df, x=x_var, y=y_var,
title=f"Relationship: {x_var} vs {y_var}")
st.plotly_chart(fig, use_container_width=True)
correlation = self.df[x_var].corr(self.df[y_var])
st.metric("Correlation", f"{correlation:.3f}")
if abs(correlation) > 0.7:
strength = "Strong"
elif abs(correlation) > 0.3:
strength = "Moderate"
else:
strength = "Weak"
direction = "positive" if correlation > 0 else "negative"
st.write(f"**Result:** {strength} {direction} correlation")
self.add_insight(f"{strength} correlation ({correlation:.3f}) between {x_var} and {y_var}", 4)
# Group analysis
if categorical_cols and numeric_cols:
st.subheader("Group Analysis")
col1, col2 = st.columns(2)
with col1:
group_var = st.selectbox("Group by:", categorical_cols)
with col2:
metric_var = st.selectbox("Analyze:", numeric_cols)
group_stats = calculate_group_stats(self.df, group_var, metric_var)
st.dataframe(group_stats, use_container_width=True)
# Sample for visualization if too many groups
unique_groups = self.df[group_var].nunique()
if unique_groups <= 20:
fig = px.box(self.df, x=group_var, y=metric_var,
title=f"{metric_var} by {group_var}")
st.plotly_chart(fig, use_container_width=True)
else:
st.info(f"Too many groups ({unique_groups}) for visualization. Showing statistics only.")
best_group = group_stats['mean'].idxmax()
best_value = group_stats.loc[best_group, 'mean']
self.add_insight(f"'{best_group}' has highest average {metric_var}: {best_value:.2f}", 4)
def stage_5_summary(self):
"""Stage 5: Summary and Export"""
st.subheader("π Analysis Summary")
# Key metrics
col1, col2, col3 = st.columns(3)
with col1:
st.metric("Total Insights", len(self.insights))
with col2:
quality = "High" if self.stats['missing_values'] == 0 else "Medium"
st.metric("Data Quality", quality)
with col3:
st.metric("Analysis Complete", "β
")
# Insights summary
st.subheader("Key Insights")
for i, insight in enumerate(self.insights, 1):
st.write(f"{i}. **Stage {insight['stage']}:** {insight['insight']}")
# Export options
st.subheader("Export Results")
export_format = st.selectbox("Choose export format:",
["Text Report", "Markdown Report", "Python Code", "Cleaned Data"])
if export_format == "Text Report":
report = self.generate_text_report()
st.download_button(
label="Download Text Report",
data=report,
file_name="analysis_report.txt",
mime="text/plain"
)
elif export_format == "Markdown Report":
report = self.generate_markdown_report()
st.download_button(
label="Download Markdown Report",
data=report,
file_name="analysis_report.md",
mime="text/markdown"
)
elif export_format == "Python Code":
code = self.generate_python_code()
st.code(code, language="python")
st.download_button(
label="Download Python Script",
data=code,
file_name="analysis_script.py",
mime="text/plain"
)
else: # Cleaned Data
# Offer different export formats
data_format = st.selectbox("Choose data format:",
["CSV", "Excel", "Parquet"])
if st.button("Export Data"):
try:
if data_format == "CSV":
csv = self.df.to_csv(index=False)
st.download_button(
label="Download CSV",
data=csv,
file_name="cleaned_data.csv",
mime="text/csv"
)
elif data_format == "Excel":
excel_buffer = BytesIO()
self.df.to_excel(excel_buffer, index=False)
excel_data = excel_buffer.getvalue()
st.download_button(
label="Download Excel",
data=excel_data,
file_name="cleaned_data.xlsx",
mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
)
else: # Parquet
parquet_buffer = BytesIO()
self.df.to_parquet(parquet_buffer, index=False)
parquet_data = parquet_buffer.getvalue()
st.download_button(
label="Download Parquet",
data=parquet_data,
file_name="cleaned_data.parquet",
mime="application/octet-stream"
)
except Exception as e:
st.error(f"Error exporting data: {str(e)}")
def generate_text_report(self) -> str:
"""Generate text analysis report"""
report = f"""DATA ANALYSIS REPORT
==================
Dataset Overview:
- Rows: {self.stats['shape'][0]:,}
- Columns: {self.stats['shape'][1]:,}
- Missing Values: {self.stats['missing_values']:,}
- Memory Usage: {self.stats['memory_usage']:.1f} MB
Key Insights:
"""
for insight in self.insights:
report += f"\n- Stage {insight['stage']}: {insight['insight']}"
report += f"\n\nGenerated: {pd.Timestamp.now().strftime('%Y-%m-%d %H:%M:%S')}"
return report
def generate_markdown_report(self) -> str:
"""Generate markdown analysis report"""
report = f"""# Data Analysis Report
## Dataset Overview
* **Rows:** {self.stats['shape'][0]:,}
* **Columns:** {self.stats['shape'][1]:,}
* **Missing Values:** {self.stats['missing_values']:,}
* **Memory Usage:** {self.stats['memory_usage']:.1f} MB
## Data Types
```
{pd.DataFrame(self.stats['dtypes'].items(), columns=['Type', 'Count']).to_markdown()}
```
## Key Insights
"""
# Group insights by stage
for stage in range(1, 6):
stage_insights = [i for i in self.insights if i['stage'] == stage]
if stage_insights:
report += f"\n### Stage {stage}\n"
for insight in stage_insights:
report += f"* {insight['insight']}\n"
report += f"\n\n*Generated: {pd.Timestamp.now().strftime('%Y-%m-%d %H:%M:%S')}*"
return report
def generate_python_code(self) -> str:
"""Generate reproducible Python code"""
code = """import pandas as pd
import numpy as np
import plotly.express as px
from typing import Dict, List, Any
# Load and prepare data
df = pd.read_csv('your_data.csv') # Update with your data source
# Basic statistics
def calculate_basic_stats(df: pd.DataFrame) -> Dict[str, Any]:
return {
'shape': df.shape,
'memory_usage': float(df.memory_usage(deep=True).sum() / 1024**2),
'missing_values': int(df.isnull().sum().sum()),
'dtypes': df.dtypes.value_counts().to_dict(),
'duplicates': int(df.duplicated().sum())
}
stats = calculate_basic_stats(df)
print("\\nBasic Statistics:")
print(f"- Shape: {stats['shape']}")
print(f"- Memory Usage: {stats['memory_usage']:.1f} MB")
print(f"- Missing Values: {stats['missing_values']}")
print(f"- Duplicates: {stats['duplicates']}")
"""
# Add data cleaning operations if any were performed
if hasattr(self, 'cleaning_history'):
code += "\n# Data Cleaning\n"
for operation in self.cleaning_history:
if "missing values" in operation.lower():
code += "# Handle missing values\n"
code += "df = df.fillna(method='ffill') # Update with your chosen method\n"
elif "duplicate" in operation.lower():
code += "# Remove duplicates\n"
code += "df = df.drop_duplicates()\n"
elif "outlier" in operation.lower():
code += """# Handle outliers
def remove_outliers(df: pd.DataFrame, column: str) -> pd.DataFrame:
Q1 = df[column].quantile(0.25)
Q3 = df[column].quantile(0.75)
IQR = Q3 - Q1
return df[~((df[column] < (Q1 - 1.5 * IQR)) | (df[column] > (Q3 + 1.5 * IQR)))]
# Apply to numeric columns as needed
numeric_cols = df.select_dtypes(include=[np.number]).columns
for col in numeric_cols:
df = remove_outliers(df, col)
"""
# Add visualization code
code += """
# Visualizations
def plot_missing_values(df: pd.DataFrame):
missing = df.isnull().sum()
if missing.sum() > 0:
missing = missing[missing > 0]
fig = px.bar(x=missing.index, y=missing.values,
title='Missing Values by Column')
fig.show()
def plot_correlations(df: pd.DataFrame):
numeric_cols = df.select_dtypes(include=[np.number]).columns
if len(numeric_cols) > 1:
corr = df[numeric_cols].corr()
fig = px.imshow(corr, title='Correlation Matrix')
fig.show()
# Generate plots
plot_missing_values(df)
plot_correlations(df)
"""
return code |