Validate your data before you let an LLM near it
Why RAG projects fail on data quality, not model choice: a validation gate for what you feed a model, and why retrieving more can make it worse.
...blended rule set that combines EMA crosses, RSI thresholds, open-interest shifts, opening range breakout (ORB) checks, and MACD confirmation. Whenever a rule set is met, the system must push a buy or sell signal along with a short one-line explanation of which conditions triggered it. Preferred flow • Fetch live tick or minute data through Kite Connect. • Run indicator calculations in real time (pandas / ta-lib are fine). • On a trigger, send the alert—Telegram, Slack, or even console/e-mail is acceptable; I’m open to your recommendation. Code requirements • Written in clean, well-commented Python 3. • Modular functions so I can tweak parameters (periods, thresholds) without diving into core logic. • Robust error handling fo...
Automated Financial Reconcilia...identified discrepancies. Users should be able to export the final reconciliation results into Excel or CSV format. Technical Requirements: Python, Pandas, NumPy, Excel/CSV processing, transaction matching algorithms, data validation, and desktop application development using PySide6 or Tkinter. Deliverables: A functional desktop reconciliation application, multi-file data import, automated transaction matching, duplicate and discrepancy detection, manual review functionality, reconciliation summary reports, Excel/CSV exports, testing, and setup documentation. Suggested tags you need to select: - Python - Financial Analysis - Data Processing - Pandas - Excel - Data Reconciliation - Desktop Application - Data Validation - Automation - Sof...
I’m putting together a beginner-friendly machine-learning demo that predicts student test scores. I’d like a clean, well-commented Python solution that a newcomer can open, run, and learn from without feeling overwhelmed. Here is what I have in mind: • A small, publicly available CSV (or a synthetic set you generate) containing study hours, attendance, or similar features, loaded with pandas. • Exploratory analysis that highlights patterns in plain language. • A straightforward regression pipeline—think scikit-learn’s LinearRegression or RandomForestRegressor—wrapped in a Jupyter notebook so each step is visible. • Clear markdown explanations of why each step matters, from data split to model evaluation (R², MAE, or R...
...an Excel file that contains: • a column for the product title • a column for the price (numeric, without currency symbols) • a column that lists the filenames or direct links of every downloaded main image Images themselves must be saved locally in a clearly named folder so I can review or upload them elsewhere. I’m flexible on the technology you choose—Python (BeautifulSoup, Selenium, pandas, openpyxl), Node, or any language you prefer—as long as the final package runs on Windows without complicated setup. Deliverables • Executable tool or script with clear usage instructions • Sample Excel file produced from at least one test link to show correct formatting • Source code, well-commented, so I can tweak or extend it...
...machine-learning mentor to steer me through the full pipeline. The goal is to create a Python/XGBoost model that can simultaneously deliver three perspectives on risk: • the chance that income falls below a chosen amount • the expected drop in income over a future period • an overall risk label (low / medium / high) To achieve this, I am looking for someone with expertise in the following areas: 1. Python, pandas, scikit-learn, XGBoost, time series analysis, handling imbalanced data, and robust model evaluation. 2. Finding relevant datasets or generating synthetic data where gaps exist—for a specific country/region. 3. Building retrieval-based AI assistants capable of providing answers grounded in user-provided data. 4. Full-stack web development with skil...
...which model produces the most reliable lift comparing accuracy, precision, recall, F1-score, confusion matrix, and ROC curves. Once the champion model is chosen, wrap it in a small Flask API for real-time scoring. A simple React dashboard should display the theft probability, key feature contributions and allow me to filter by account, date range or location. Please keep everything in Python (pandas, scikit-learn, TensorFlow/PyTorch—whichever you prefer) and make all queries database-agnostic so they run cleanly against PostgreSQL. The final repo should include: • Well-commented notebooks that reproduce the exploratory analysis, feature engineering and model training • The trained model artefacts plus a script to retrain with fresh data • Flask service with...
...levels and audience growth. The agent should automatically pull raw figures from sources such as Google Analytics, Facebook, Instagram, X/Twitter and YouTube, tidy them, then turn the numbers into easy-to-read visuals and concise written insights. Think daily or weekly runs, a quick comparison with previous periods, and a highlight reel of what changed and why. Feel free to work in Python with pandas, scikit-learn, LangChain or a similar stack, so long as the final solution can live on my VPS inside a container. Deliverables • Fully commented source code and environment file • Step-by-step setup guide • One sample report covering the last 30 days of data • Brief doc on how to plug in new data sources Acceptance criteria • End-to-end auto...
...outlier is found the program should build a clear, time-stamped report (CSV, PDF, or Markdown—whichever you prefer) summarising every unusual reading, the variable affected, its deviation from expected behaviour, and the file it came from. No live streaming or dashboards are required; the focus is on offline analysis and accurate reporting. You are free to choose the most appropriate libraries—pandas for parsing, scikit-learn, PyOD, or similar for the anomaly model, plus matplotlib/plotly if visual context strengthens the report. I value clean, well-documented code and a short README that shows how to: 1. Point the script at a directory of JSON/text logs 2. Train or load the anomaly model 3. Generate the final report in a single command The job is complete wh...
...insights, visualisation and dashboard building • Python and SQL for data cleaning, exploration and deeper analysis You’ll design and deliver interactive sessions that balance clear explanation with plenty of hands-on practice. By the end, every participant should confidently: 1. Build reports and live dashboards in Excel or Power BI. 2. Import, clean and analyse data using Python (Jupyter, pandas) and basic SQL queries. 3. Work through an end-to-end analytics project, from problem statement to presentation of findings. Please outline your proposed schedule, sample exercise ideas, and any success stories of taking complete beginners to this level of competence. I’m looking for engaging teaching style, real-world examples, and materials the students can ke...
...rounds for a chosen date range. • It calculates basic metrics such as total rounds, win / loss counts, average payout, and any other easily derived figures that help me spot patterns. • It exports everything into a single report file (CSV or XLSX preferred; PDF is a nice bonus) that I can download directly after each run. If Python makes sense, feel free to rely on requests, BeautifulSoup, pandas or similar libraries; another language is fine if it delivers the same result just as simply. Please keep the code well-commented and include a quick “how to run” note so I can repeat the process on my own server. Deliverables: 1. Fully working, documented script. 2. One sample report generated from recent Apple of Fortune data to prove the stats are accur...
...generation. - Timeline visualization. - Scenario comparisons. - PDF report generation. F. Futuristic Dashboard Create a dashboard with: - A glowing timeline. - Animated time-travel effects. - Historical and future event cards. - Interactive charts. - Before-and-after scenario comparisons. - AI-generated explanations. 5. Technology Stack - Python - Machine learning: Scikit-learn - Data processing: Pandas and NumPy - Visualization: Plotly - Interface: Streamlit - AI models: Suitable language models and predictive models - Automation: Python scheduling and workflow tools - Database: SQLite 6. Expected Output Build a functional prototype that allows users to select a year, explore historical information, modify hypothetical conditions, generate simulated future scenarios, and ...
...practitioner to step in as their mentor—someone who can translate real-world know-how into clear, engaging guidance that builds a solid foundation across these three domains. The work is fully remote and organised around weekly live sessions (Zoom or similar). You would design a logical learning path, run interactive lectures or workshops, walk through hands-on demos in Python (think Jupyter, pandas, scikit-learn, TensorFlow/Keras), and introduce essential cybersecurity principles such as basic network security and OWASP Top 10 practices. Outside class time, students will rely on you for code reviews, career tips, and occasional one-to-one Q&A. Deliverables • A 10-to-12-week syllabus with learning objectives and recommended resources • Slide decks or not...
...working with a set of raw survey-based research data and need a skilled analyst to turn it into clear, actionable insights. You’ll receive the full dataset in its original format; your job is to clean it, explore key relationships, and present the findings in an easy-to-digest form that non-technical stakeholders can understand. Please apply whichever tools you’re most comfortable with—Python (pandas, NumPy, matplotlib, seaborn), R (tidyverse), or a comparable platform—as long as the final deliverables meet the criteria below. Deliverables • A cleaned, well-documented dataset (CSV or similar) • An analytic report or slide deck summarizing methodology, statistics, and main takeaways • Clear visualisations that illustrate the most impo...
...metrics (such as Cosine Similarity or Content-Based/Collaborative Filtering) to recommend similar tracks based on user preferences. User Interface: Building a clean, interactive web dashboard using Streamlit to search for tracks, view recommendations, and display visual insights. Requirements & Deliverables: Clean & Documented Code: Well-structured Python code using standard data science libraries (Pandas, Scikit-learn, etc.). Research & Resources: Assistance in compiling a list of at least 10 relevant references/resources, adhering to these guidelines: A brief description of the contribution of each resource to the project. Project Summary/Report: A short technical report summarizing the model and results. The goal is a focused, high-quality, and robust Data Sc...
I’m building a backtesting engine and now need a robus...naming and UTC timestamps so I can drop it straight into my backtester. • Clear documentation explaining the data sources, any rate-limit handling, and how to extend the scraper when new tickers or markets are added. • A brief demo run showing sample results for one equity, one crypto pair, and one futures contract. Please use well-supported libraries (requests, BeautifulSoup, Selenium, Scrapy, pandas, etc.) and keep external dependencies minimal. Error handling, retries, and polite throttling are important—this must run unattended without breaking site rules. I’ll review the code for readability, rerun the demo, and validate that fields line up across datasets. Once everything passes those ch...
...date and text formats, handling of missing values, and correction of common formatting inconsistencies. It should also provide a comparison between the original and cleaned datasets, showing the changes made and the remaining issues. Users should be able to export the cleaned dataset in Excel or CSV format and download a summary report of the validation results. Technical Requirements Python, Pandas, AI/LLM API integration, data validation, data-cleaning techniques, desktop application development using PySide6 or Tkinter, and Excel/CSV processing. Deliverables A functional desktop application, Excel and CSV import support, automated data-quality inspection, AI-generated recommendations, data-cleaning functionality, review and approval of suggested corrections, before-and-af...
...project. I will fund an initial milestone of US $750 to get the core system designed, coded, back-tested, and ready for a small live trial. Once the strategy proves profitable and I begin generating returns, the remaining compensation will be released in full—so the sooner the bot performs, the sooner you are paid in full. Key expectations • Written in Python using robust libraries such as pandas, NumPy, ta-lib, Backtrader (or a comparable framework) • Connection to a live brokerage API (Interactive Brokers, Alpaca, or another equities gateway we agree upon) • Configurable intraday strategy engine: signal generation, position sizing, risk and money management, fail-safes, and logging • Comprehensive back-testing report covering at least two...
...comfortable with—so long as it supports programmatic order placement and basic account management. I already have the trading concept drafted, complete with entry, exit, and risk rules. What I lack is the technical bridge from concept to production code: robust Python architecture, reliable data handling, broker integration, and the testing framework to prove the idea’s merit. Libraries such as pandas, ta-lib, backtrader, or any tools you prefer are welcome as long as the final solution is readable and maintainable. Is it possible for someone to build it for me completely free of charge, without any financial compensation? Acceptance criteria for our first milestone • A standalone Python script or package that connects to the chosen stock broker’s API ...
...and attendance, past test scores with homework results, plus broader study-habit indicators such as sleep patterns and class participation. What I need from you is a small but realistic workflow that someone new to AI/ML can clone, read, and run without frustration. Think of it as a guided tutorial packaged with production-worthy code quality. Core tech I expect you to use: • Python 3.x with pandas and NumPy for data wrangling • scikit-learn for the regression pipeline (LinearRegression, RandomForestRegressor or any comparable beginner-friendly algorithms) • Jupyter Notebook (or a .py script with CLI flags) so the user can walk through each step visually Optional but welcome: matplotlib or seaborn for quick exploratory plots, and joblib or pickle for savi...
I have an assignment that centres strictly on data cleaning and producing clear, insightful charts. The raw file is already with me in Excel/CSV format, so you can work directly in Excel itself or use Python (pandas, matplotlib, seaborn) or R if you prefer, as long as the final assets are easy for me to open in Excel. Scope • Clean the dataset: detect and fix inconsistencies, handle missing or duplicate rows, standardise column names and formats. • Create meaningful visualisations: bar, line, pie, or any chart type that best highlights the main trends you uncover. All charts must sit on a separate sheet or be delivered as image files ready to paste into a report. Deliverables 1. A fully cleaned Excel or CSV file replacing the original. 2. A set of polished charts...
...characters and non-English text remain intact. • Uses Pandas for parsing and OpenPyXL for writing the resulting Excel file. • Preserves every column, row, and cell value exactly as it appears in the source. • Generates a downloadable .xlsx file and confirms completion to the user. Deliverables 1. Executable desktop application for Windows (macOS/Linux packaging is a plus but not mandatory). 2. Well-structured, commented source code. 3. A brief README that explains setup, dependencies, and how to extend or customize the tool. I value efficiency, clear error handling, and a straightforward UI—ideally a single-window interface with a “Select CSV” button and a “Convert” action. If you have prior experience combining Pa...
...• Applies thorough cleaning: deduplication, missing-value handling, type standardisation, outlier checks, and consistent naming so that the dataset meets common ML requirements. • Loads the refined data back into a staging database or delivers it as flat files (CSV, Parquet, etc.) ready to drop into my analytics notebooks. I’m comfortable working in Python, so solutions that leverage Python, Pandas, SQL Alchemy, or even an SSIS package are all viable as long as you supply every script, configuration file, and a concise runbook explaining how to rerun the process end-to-end. Acceptance criteria 1. No duplicate primary keys or orphaned references remain in the cleaned tables. 2. Null rates are reduced to an agreed tolerance and documented. 3. The entire pi...
...programming-focused college assignment that sits squarely in the data analysis / machine-learning space. The core requirement is to take a supplied dataset, clean it thoroughly, run the appropriate exploratory and statistical analyses, and extract clear, defensible insights that I can present in class. Here’s the flow I have in mind: • Receive the raw dataset from me (CSV or Excel). • Use Python—pandas, NumPy, and, where helpful, scikit-learn or SciPy—to handle missing values, correct data types, remove outliers, and generally prepare the data for analysis. • Perform exploratory data analysis: descriptive statistics, correlations, and any hypothesis tests that reveal meaningful patterns. • Summarise findings in both code comments and ...
...and customer records and I need a clear, data-driven picture of which products and services actually drive revenue. Your task is to take that raw information, clean and consolidate it, then surface an undeniable ranking of our best-sellers. You will be working with sales and customer data only, with the primary objective of finding the best-selling products or services. Feel free to use Python (pandas), Excel Power Query, Power BI, Tableau or any tool you prefer, as long as the insights are easy for non-technical stakeholders to understand. Deliverables: • A ranked list of top-performing products/services with supporting metrics (units sold, revenue, margin if calculable). • A concise visual report or dashboard so the leadership team can slice results by date range,...
I have two separate customer-contact spreadsheets that no longer line up. Your task is to compare them side-by-side, using Full Name as the sole matching key, and clearly identify every contact that exists in one list but not in the other. Accuracy is crucial because these files feed different parts of our sales cycle. You’re free to work in Excel, Google Sheets, Python (pandas), or any similar tool—whatever lets you deliver a clean, verifiable result fastest. I’ll supply both raw files as soon as we start. Deliverables should arrive as: • A new sheet (or CSV) containing only the missing contacts, flagged by which original list they were absent from • A brief note on any name-format inconsistencies you had to normalise so I can replicate the process...
...You will receive the raw text file (CSV). I expect back the same file structure, duplicates removed, and a short log outlining what was eliminated so I can trace the changes. A quick turnaround and careful attention to detail are important, as any missed duplicate will skew later processing. If you’re comfortable working with large text datasets in tools such as Excel, Google Sheets, or Python (pandas), you’ll find this straightforward....
...employs—without breaking current functionality. 4. Log actions and flag mismatches (e.g., EAN not found) to a timestamped report. 5. Provide the updated script and a short README explaining any new config options or dependencies. You will have access to the current script, a sample supplier file, and test credentials for a staging shop. If you know Gambio internals and are comfortable with pandas, openpyxl, or similar Python data-handling libraries, this should be a quick enhancement....
...technical problem-solving abilities. • Ability to collaborate effectively with multidisciplinary researchers. • U.S. citizenship is required due to government access requirements. Preferred Qualifications • Experience with probabilistic modeling, Bayesian methods, or uncertainty quantification. • Experience with Monte Carlo simulation and synthetic data generation. • Familiarity with NumPy, SciPy, pandas, scikit-learn, or comparable scientific computing tools. • Experience implementing research algorithms from mathematical specifications. • Knowledge of scientific software verification and validation. • Experience designing secure APIs or modular software integration interfaces. • Prior involvement in federally funded research or scie...
I’ve gathered a series of CSV files that record every sale we’ve made and now need a concise Python-driven study that pinpoints which products bring in the highest revenue. The data set is strictly sales data, and the key metric for ranking each item is revenue generated, not sheer volume or frequency. You’ll write a clean, well-commented Python script (pandas / NumPy are fine; a Jupyter Notebook is equally acceptable) that: • reads one or many CSVs, • calculates total revenue for each SKU, and • outputs a tidy summary—both as a CSV and a readable table or chart—listing products in descending order of revenue. Acceptance criteria • Script runs locally with a single command and works on standard Python 3.10+ • All file...
I have a raw dataset that first needs a clear descriptive analysis so I can understand the basic patterns hiding inside it. Once those insights a...written summary of the descriptive statistics (counts, means, medians, standard deviation, common ranges—whatever best fits the data). • Clean, well-labeled PNG graphics that visually communicate those trends. Think bar charts, line graphs, or any plot type that makes the story pop. • All code or spreadsheets used, so I can reproduce or extend the work later. If you normally work in Python (Pandas, Matplotlib, Seaborn), R (ggplot2), or a spreadsheet tool, that’s fine—just keep the workflow tidy and documented. Deliver once the visuals line up perfectly with the insights; if something looks unclear I&rsqu...
...report that a non-technical stakeholder can understand at a glance. The scope centres on efficiency only—speed, durability and other metrics are not required at this stage—so every test, instrument choice and data point should directly support that single KPI. You are free to propose the most suitable tools (e.g., high-precision power meters, data loggers, statistical software such as R, Python/pandas) provided they produce reproducible results. Deliverables • A brief test plan outlining methodology, instrumentation and acceptance limits • The complete, unaltered data set in CSV or XLSX format • A summary report highlighting efficiency findings, key observations and any actionable insights Acceptance criteria • Test plan approved before ...
...I don’t have any tick or candle source to work with, so part of the job is to propose and hook up a reliable data feed that includes historical weekend prices and enough depth to calculate gaps back-testing style. Free solutions are welcome if they are stable; paid APIs are fine too as long as the cost is clear up-front. Once the data is flowing I want clean, well-commented code—Python with pandas or an equivalent stack is my first choice, but I’m open to whatever language you feel produces the most transparent results. The output I expect for each pair is: • size of the weekend gap (in pips) • maximum adverse movement measured in pips before the gap is filled • a timestamp so I can trace any individual event Please package the finding...
...data-quality checks (shape, duplicated rows, obvious null counts) completed. From here, I need to take the work through to a submission-ready assignment that meets the brief and marking rubric supplied by my course. Scope of work • Polish the preprocessing pipeline: tackle any remaining missing-value issues, tidy categorical variables, and prepare features so they are ready for modeling both in Python (Pandas, scikit-learn) and, if easier for you, in WEKA (ARFF export). • Carry out three clearly defined lines of analysis that shed light on casualty severity. You are free to choose the angles provided they are relevant, insightful, and defensible in the report. • Build and evaluate at least one suitable predictive or explanatory model (for example logistic re...
I’m working with a CSV dataset that mixes numerical and categorical fields and need a complete, well-documented Jupyter Notebook delivered within the next four days. The notebook should walk through data cleaning and exploration with Pandas and NumPy, focus on uncovering trends in the data, visualise key findings through clear charts or graphs, and finish with a straightforward machine-learning model built in scikit-learn. Please structure your code so each step—loading, preprocessing, exploratory analysis, visualisation, modelling, and interpretation—is easy to follow. Inline markdown explanations are essential because I intend to use the notebook as both a reference and a learning resource. Deliverables: • Jupyter Notebook (.ipynb) with commented, reprod...
...practitioner to step in as their mentor—someone who can translate real-world know-how into clear, engaging guidance that builds a solid foundation across these three domains. The work is fully remote and organised around weekly live sessions (Zoom or similar). You would design a logical learning path, run interactive lectures or workshops, walk through hands-on demos in Python (think Jupyter, pandas, scikit-learn, TensorFlow/Keras), and introduce essential cybersecurity principles such as basic network security and OWASP Top 10 practices. Outside class time, students will rely on you for code reviews, career tips, and occasional one-to-one Q&A. Deliverables • A 10-to-12-week syllabus with learning objectives and recommended resources • Slide decks or not...
...Document every step so another analyst could reproduce your results exactly Critical skills - Survey methodology or psychometrics: scale design, reliability and validity testing; item response theory or Rasch modelling a strong plus - Applied statistics: sampling design, confidence intervals, estimating from a sample to a whole population - Python or R for data cleaning, analysis and automation (pandas, tidyverse or similar) - Hands-on experience using large language models to code or classify text, and checking that work against human judgment - Comfortable with messy real-world data from spreadsheets and business-software exports - Clear written English for reports and documentation The person we want - Rigorous and sceptical: you check your own numbers before anyone else doe...
I’m looking for a junior-level data scientist who is comfortable working end-to-end in Python and its key libraries (Pandas, NumPy, scikit-learn, Matplotlib, Seaborn, etc.). The work will cover three main areas: • Data cleaning and preparation • Exploratory analysis with clear, insightful visualizations • Building and evaluating predictive models that can be handed off for production use You’ll receive raw datasets along with a brief describing the business questions. I’ll need you to: 1. Inspect and clean the data, documenting every assumption and transformation. 2. Produce visual summaries that highlight trends, anomalies, and any correlations worth noting. 3. Develop at least one baseline predictive model, iterate to improve performan...
...engagement metrics—that need to be explored and transformed into clear, actionable insights aimed squarely at boosting retention. What I’m looking for is a concise, end-to-end analysis that not only quantifies churn and loyalty drivers but also surfaces the patterns that explain why our best users stay and others leave. Here’s what I need from you: • A clean, well-documented workflow (Python, SQL, pandas preferred) that ingests the raw data, handles missing values, and engineers any retention-focused features. • Exploratory analysis that highlights cohort trends, lifecycle touchpoints, and at-risk segments, backed by visualizations in Tableau or a similar BI tool. • A succinct report—slides or notebook—summarizing key findings, ...
I’m looking to replace a daily copy-and-paste routine with a hands-free workflow that pipes records from our existing databases straight into the bookkeeping platform. The goal is simple: full...• Well-commented code (Python, SQL scripts, or another mainstream language you prefer) plus a step-by-step deployment guide I’ll provide test credentials, schema diagrams, and the API documentation. Once everything runs smoothly for a week in staging—no failed jobs, no mismatched balances—the project will be considered complete. When you respond, outline the approach, language, and libraries (pandas, SQLAlchemy, airflow, etc.) you’d lean on, and point me to any similar automation you’ve delivered. Let’s make data entry fully automatic so the f...
...scripts re-run byte for byte, no tolerance in the code, every reported figure in a script output. 3. Exact reproduction - Rebuild at least one real, independently audited case per engine from its published documents: every intermediate and final figure exact, zero tolerance. If a case cannot be rebuilt, the engine is not kept. - Exact decimal arithmetic throughout; no binary floating point, so no pandas or NumPy for calculations. - Round and truncate where and how the documents do, line by line or on totals, recording the rule and the reason beside each rounded figure; the rule is found in the body's texts, never assumed. - No difference is accepted until four questions are answered with arithmetic: same basis and period? same unit and scale? derived by a stated rule? expla...
...noise often found in host names, neighbourhood strings, and price columns. Next, handle missing values thoughtfully—no blanket deletions—so the final table truly represents the market. Once the data are tidy, I want a descriptive analysis that answers, at minimum, “What has the average nightly price looked like over time, and how does it differ by location and listing category?” Python with pandas is my preferred toolkit, but R (tidyverse) is fine if that’s your strength. Visualisations created in Matplotlib, Seaborn, or ggplot2 should accompany the summary so the insights are instantly readable. Deliverables • Cleaned master Excel/CSV file ready for further modelling • Reproducible script or Jupyter Notebook with well-commented c...
...exactly as they are. Second, create a separate CSV that contains a random, evenly-distributed sample of exactly 230 records for every Unit. The sample has to consider every Road and MapPage so the distribution remains balanced, and the final file must be sorted alphabetically by Unit, Road, then MapPage. Please handle the heavy lifting with a more efficient approach than Excel or Access—Python (pandas), SQL, or any equivalent tool is fine with me as long as it runs quickly on a dataset this size. The source CSV is through whichever transfer method you prefer once we start. Deliverables • Rolled-up CSV as described above • Random-selection CSV with 230 balanced records per Unit • Reproducible script or query so I can rerun the process later Compl...
I keep a collection of text-only records in several CSV files and need a solid validation pass before the next release cycle. The task is to build and run checks that confirm each field adheres to agreed formats, catches missing or duplicate entries, flags unexpected characters or line breaks, and produces a concise report of every issue found. You are free to use Python (pandas, Great Expectations, or similar), Bash + awk, SQL import routines—whatever gets reliable, repeatable results. Everything must run headless on Ubuntu 22.04 and be callable from the command line so it can slip into our existing CI pipeline. Deliverables: • A runnable script or notebook with clear setup instructions • A summary report (CSV or Markdown) listing all validation errors and the...
...- Develop Python scripts using Pandas for data mapping and transformation. - Map source columns to target columns based on mapping rules. - Handle missing values, duplicate records, data type conversions, and invalid data. - Perform data cleansing, validation, and error handling. - Support CSV, Excel, JSON, and SQL data sources. - Generate transformed output files and validation reports. - Write reusable, well-documented, and maintainable code. **Expected Deliverables:** - Complete Python source code. - Configurable source-to-target mapping rules. - Sample input and output files. - Data validation and error-handling logic. - Setup instructions and documentation. Please share your approach, estimated delivery time, and relevant experience with Python, Pandas, SQL, and ETL...
...milestones are: • Build a repeatable mapping routine that pulls source tables from SQL Server, aligns every field to a target schema, and flags any unmapped or ambiguous values for review. • Implement allocation logic so amounts, quantities or percentages are distributed across multiple target records according to business rules I will supply. Everything should run in Python 3.x, ideally using pandas and SQL-Alchemy (or pyodbc) to keep the workflow fast and transparent. Clear comments, minimal external dependencies, and a short README are part of the hand-off. I will verify the work by running your scripts against a staging database; all mapped fields must reconcile 100 % and allocation totals must balance to the source. If you have solid experience combining P...
I have several datasets coming in from different systems where the numeric fields are stored as plain ASCII files in some cases and UTF-8 exports in others. My goal is to run solid, reproducible data analysis on them, but first they need to be cleaned and standardised. Here is what I need from you: • Build a repeatable cleaning pipeline in Python (Pandas, NumPy) that detects the file encoding, converts everything to UTF-8 where necessary, and validates that all numeric columns are truly numeric—no stray characters, misplaced separators, or locale-specific quirks. • Where the raw files already sit in SQL Server, create T-SQL or Python-based routines that achieve the same checks and corrections in-database so I can schedule them in an Agent job. • Document...
I ha...captured accurately and placed into clearly-labelled columns in Excel—no images, just usable text values. 2. From one public website, I only need the visible text content (no images or charts). The pages are standard HTML. Pull that text, keep its section headings intact, and add it to separate sheets in the same workbook so I can cross-reference it later. Please automate both steps—Python with pandas, Beautiful Soup, or any reliable toolset is fine—as long as I receive: • A single XLSX file with separate sheets for: – each PDF form dataset – each scraped web page’s text • The well-commented script(s) used, so I can rerun the process when new PDFs or pages appear. Accuracy takes priority over speed; I will spot-check...
...text cleaning and TF‑IDF Vectorization to turn words into numbers. Ran the Multinomial Naive Bayes model to predict labels. Separated the data into training and test parts to check performance. Checked outcomes with accuracy, precision, recall and F1‑score. Made a Streamlit app so you can try your own messages. Loaded the model and TF‑IDF tool with Joblib. Technologies Used Python | NLP | Pandas | Scikit‑learn TF‑IDF | Multinomial Naive Bayes | Streamlit Joblib Dataset SMS Spam Collection Dataset. UCI Machine Learning Repository The model learns from this labelled data. Its guesses rely on the patterns from this set. Might not always be right, on brand‑new messages. Project Outcome I produced a ready‑to‑use text classifier that shows how NLP and machine learning can spot ...
...From MGT-7: current shareholding pattern and the key financial figures filed for the year • From AOC-4: the latest company profile information together with the list of directors The final deliverable should be an Excel or CSV file where each company occupies one row and the above data points sit in clearly labelled columns. If you would like to automate the process, feel free to use Python, pandas, Beautiful Soup, Selenium, or any other stack you are comfortable with—just ensure the script can be rerun should I need an update later. Accuracy is critical; every figure must match the original MCA document. I will review a small sample first, and once verified we can move on to the full extraction. Let me know your proposed approach and the timeframe you need....
Deliver production-leaning Python: readable names, small functions, and a requirements file if needed. Include a `if __name__ == '__main__'` entry point and one documented example command. The Python work centers on: Jupyter notebook to plain script refactor. Synthetic/sample inputs are enough for the demo run—no live production keys required. Tech stack: Python 3 only. Reference batch ID: QA-PY-0148.
Why RAG projects fail on data quality, not model choice: a validation gate for what you feed a model, and why retrieving more can make it worse.
Learn how Python web scraping can turn raw website data into clean, structured datasets ready for analysis, reporting, and business insights.
Is it the end of an era? Is there any point in learning how to be a programmer if it's going to get taken over in the future?