Data Science Practice Questions: Roadmap-Wise Sheets

int, float, complex, bool, str, list, tuple, set, dict and NoneType. Mutable ones are list, set and dict; the others are immutable. Example application: In a hypothetical ServiceNow project involving IT service management and enterprise workflow automation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

List: ordered and mutable. Tuple: ordered and immutable (hashable, slightly faster). Set: unordered, unique items, fast membership checks. Dict: key–value pairs with fast lookup by key. Example application: In a hypothetical HCLTech project involving enterprise technology services and infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

An iterable can produce an iterator (it has __iter__); an iterator has __next__ and remembers its position. A for-loop calls iter() on the iterable and then next() repeatedly. Example application: In a hypothetical TCS project involving enterprise IT services, integration, and client delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

An iterable can produce an iterator (it has __iter__); an iterator has __next__ and remembers its position. A for-loop calls iter() on the iterable and then next() repeatedly. Example application: In a hypothetical Uber project involving ride-hailing, driver operations, and real-time dispatch, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks. Example application: In a hypothetical Stripe project involving online payments, billing, and financial infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks. Example application: In a hypothetical CRED project involving member rewards, credit-card payments, and financial products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks. Example application: In a hypothetical ServiceNow project involving IT service management and enterprise workflow automation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

An iterable can produce an iterator (it has __iter__); an iterator has __next__ and remembers its position. A for-loop calls iter() on the iterable and then next() repeatedly. Example application: In a hypothetical Goldman Sachs project involving investment banking, markets, and risk analytics, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

An iterable can produce an iterator (it has __iter__); an iterator has __next__ and remembers its position. A for-loop calls iter() on the iterable and then next() repeatedly. Example application: In a hypothetical Cisco project involving networking, security, and collaboration infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

An iterable can produce an iterator (it has __iter__); an iterator has __next__ and remembers its position. A for-loop calls iter() on the iterable and then next() repeatedly. Example application: In a hypothetical Barclays project involving retail banking, trading, and financial compliance, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A decorator wraps a function to add behaviour such as logging, timing or caching without changing its code. It is applied with the @name syntax above the function. Example application: In a hypothetical Tech Mahindra project involving telecom platforms, IT services, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A decorator wraps a function to add behaviour such as logging, timing or caching without changing its code. It is applied with the @name syntax above the function.

== compares values for equality, while is checks whether two names point to the same object in memory. Use is mainly for None checks. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The rule that lets arrays of different shapes operate together by virtually stretching the smaller one, for example adding a scalar or a (3,) vector to a (4,3) matrix. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

def rmse(y, p): return (sum((a - b) ** 2 for a, b in zip(y, p)) / len(y)) ** 0.5 Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

loc selects by label (row and column names, inclusive of the end label); iloc selects by integer position (end-exclusive). Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

List: ordered and mutable. Tuple: ordered and immutable (hashable, slightly faster). Set: unordered, unique items, fast membership checks. Dict: key–value pairs with fast lookup by key. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

int, float, complex, bool, str, list, tuple, set, dict and NoneType. Mutable ones are list, set and dict; the others are immutable. Example application: In a hypothetical Cisco project involving networking, security, and collaboration infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A function using yield that produces values lazily, one at a time. It saves memory when processing large files or streams because the full sequence is never held in RAM. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A mutex in CPython that lets only one thread execute Python bytecode at a time. Use multiprocessing for CPU-bound work and threads or asyncio for I/O-bound work. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A decorator wraps a function to add behaviour such as logging, timing or caching without changing its code. It is applied with the @name syntax above the function. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

df.groupby('region').agg(total=('sales','sum'), avg=('sales','mean')) gives named aggregations per group in one pass. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

(df.assign(rev=df.price * df.qty) .groupby(['region', 'product'])['rev'].sum() .reset_index() .sort_values('rev', ascending=False) .groupby('region').head(3)) Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A method is a function that belongs to an object or class and receives the instance (self) as its first argument. Example application: In a hypothetical LinkedIn project involving professional networking, job discovery, and recruiting products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

def rmse(y, p): return (sum((a - b) ** 2 for a, b in zip(y, p)) / len(y)) ** 0.5 Example application: In a hypothetical Stripe project involving online payments, billing, and financial infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A method is a function that belongs to an object or class and receives the instance (self) as its first argument. Example application: In a hypothetical Swiggy project involving food delivery, logistics, and quick commerce, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A method is a function that belongs to an object or class and receives the instance (self) as its first argument. Example application: In a hypothetical ServiceNow project involving IT service management and enterprise workflow automation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

def rmse(y, p): return (sum((a - b) ** 2 for a, b in zip(y, p)) / len(y)) ** 0.5 Example application: In a hypothetical Walmart project involving retail commerce, inventory, and supply-chain operations, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

def rmse(y, p): return (sum((a - b) ** 2 for a, b in zip(y, p)) / len(y)) ** 0.5 Example application: In a hypothetical JP Morgan Chase project involving banking, payments, and financial risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A method is a function that belongs to an object or class and receives the instance (self) as its first argument. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

def rmse(y, p): return (sum((a - b) ** 2 for a, b in zip(y, p)) / len(y)) ** 0.5 Example application: In a hypothetical Cisco project involving networking, security, and collaboration infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

def rmse(y, p): return (sum((a - b) ** 2 for a, b in zip(y, p)) / len(y)) ** 0.5

A method is a function that belongs to an object or class and receives the instance (self) as its first argument.

Small anonymous one-line functions, handy as arguments to sort, map, filter or Pandas apply. Example application: In a hypothetical D. E. Shaw project involving quantitative research and financial technology, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Small anonymous one-line functions, handy as arguments to sort, map, filter or Pandas apply. Example application: In a hypothetical Razorpay project involving payment gateway, settlement, and merchant integrations, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The rule that lets arrays of different shapes operate together by virtually stretching the smaller one, for example adding a scalar or a (3,) vector to a (4,3) matrix. Example application: In a hypothetical Amazon project involving an online marketplace, fulfilment operations, and cloud services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The rule that lets arrays of different shapes operate together by virtually stretching the smaller one, for example adding a scalar or a (3,) vector to a (4,3) matrix. Example application: In a hypothetical Adobe project involving creative software, document workflows, and subscriptions, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The rule that lets arrays of different shapes operate together by virtually stretching the smaller one, for example adding a scalar or a (3,) vector to a (4,3) matrix. Example application: In a hypothetical Intuit project involving financial software, tax preparation, and small-business tools, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks. Example application: In a hypothetical Capgemini project involving client delivery, data platforms, and enterprise transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks. Example application: In a hypothetical Amazon project involving an online marketplace, fulfilment operations, and cloud services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks. Example application: In a hypothetical Microsoft project involving developer tools, cloud services, and productivity applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks. Example application: In a hypothetical PayPal project involving digital payments, merchant checkout, and transaction risk, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks. Example application: In a hypothetical Meesho project involving social commerce, seller listings, and logistics, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks. Example application: In a hypothetical Visa project involving card networks, authorization, and payment-risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks.

They store same-type values in contiguous memory and run vectorised operations in compiled C, avoiding Python loop overhead and per-object type checks.

loc selects by label (row and column names, inclusive of the end label); iloc selects by integer position (end-exclusive). Example application: In a hypothetical Salesforce project involving CRM, sales operations, and customer data workflows, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

loc selects by label (row and column names, inclusive of the end label); iloc selects by integer position (end-exclusive). Example application: In a hypothetical Zepto project involving quick commerce, inventory, and last-mile delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

merge joins on columns, SQL-style (how='inner', 'left', ...). join joins on the index. concat stacks DataFrames along rows or columns. Example application: In a hypothetical Infosys project involving enterprise consulting, digital platforms, and client transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

(df.assign(rev=df.price * df.qty) .groupby(['region', 'product'])['rev'].sum() .reset_index() .sort_values('rev', ascending=False) .groupby('region').head(3)) Example application: In a hypothetical Zomato project involving food ordering, restaurant discovery, and delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

pd.pivot_table(df, index='region', columns='month', values='sales', aggfunc='sum') reshapes long data into a summary matrix; melt() does the reverse. Example application: In a hypothetical Swiggy project involving food delivery, logistics, and quick commerce, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

merge joins on columns, SQL-style (how='inner', 'left', ...). join joins on the index. concat stacks DataFrames along rows or columns. Example application: In a hypothetical Paytm project involving payments, merchant services, and financial products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

pd.pivot_table(df, index='region', columns='month', values='sales', aggfunc='sum') reshapes long data into a summary matrix; melt() does the reverse. Example application: In a hypothetical Morgan Stanley project involving wealth management, trading, and investment platforms, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

map works element-wise on a Series, apply works on a Series or along a DataFrame axis, and applymap (now DataFrame.map) applies a function to every cell. Example application: In a hypothetical HCLTech project involving enterprise technology services and infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

map works element-wise on a Series, apply works on a Series or along a DataFrame axis, and applymap (now DataFrame.map) applies a function to every cell.

merge joins on columns, SQL-style (how='inner', 'left', ...). join joins on the index. concat stacks DataFrames along rows or columns.

Read in chunks or use usecols and dtypes, convert strings to category, avoid apply with row loops in favour of vectorised ops, and consider Polars, Dask or Parquet storage. Example application: In a hypothetical Cognizant project involving enterprise applications and healthcare or financial client systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Read in chunks or use usecols and dtypes, convert strings to category, avoid apply with row loops in favour of vectorised ops, and consider Polars, Dask or Parquet storage. Example application: In a hypothetical ServiceNow project involving IT service management and enterprise workflow automation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level. Example application: In a hypothetical Wipro project involving managed IT services, cloud migration, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level. Example application: In a hypothetical Accenture project involving multi-client consulting, cloud transformation, and process automation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level. Example application: In a hypothetical LinkedIn project involving professional networking, job discovery, and recruiting products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level. Example application: In a hypothetical Stripe project involving online payments, billing, and financial infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level. Example application: In a hypothetical PhonePe project involving UPI payments, merchant transactions, and financial services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level. Example application: In a hypothetical Razorpay project involving payment gateway, settlement, and merchant integrations, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level. Example application: In a hypothetical Swiggy project involving food delivery, logistics, and quick commerce, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level. Example application: In a hypothetical Paytm project involving payments, merchant services, and financial products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level. Example application: In a hypothetical HCLTech project involving enterprise technology services and infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level. Example application: In a hypothetical Tech Mahindra project involving telecom platforms, IT services, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level.

Power is the probability of detecting a real effect. Increase it with a larger sample, a bigger effect size, lower noise or a higher significance level.

Modelling the count of events in a fixed interval when events occur independently at a constant average rate, such as calls per hour or defects per batch. Example application: In a hypothetical TCS project involving enterprise IT services, integration, and client delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A symmetric bell-shaped distribution defined by mean and standard deviation, where about 68%, 95% and 99.7% of values lie within 1, 2 and 3 standard deviations. Example application: In a hypothetical Amazon project involving an online marketplace, fulfilment operations, and cloud services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A symmetric bell-shaped distribution defined by mean and standard deviation, where about 68%, 95% and 99.7% of values lie within 1, 2 and 3 standard deviations. Example application: In a hypothetical Microsoft project involving developer tools, cloud services, and productivity applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Modelling the count of events in a fixed interval when events occur independently at a constant average rate, such as calls per hour or defects per batch. Example application: In a hypothetical LinkedIn project involving professional networking, job discovery, and recruiting products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Modelling the count of events in a fixed interval when events occur independently at a constant average rate, such as calls per hour or defects per batch. Example application: In a hypothetical Bloomberg project involving financial terminals, market data, and news systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A symmetric bell-shaped distribution defined by mean and standard deviation, where about 68%, 95% and 99.7% of values lie within 1, 2 and 3 standard deviations. Example application: In a hypothetical PayPal project involving digital payments, merchant checkout, and transaction risk, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Modelling the count of events in a fixed interval when events occur independently at a constant average rate, such as calls per hour or defects per batch. Example application: In a hypothetical Meesho project involving social commerce, seller listings, and logistics, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Modelling the count of events in a fixed interval when events occur independently at a constant average rate, such as calls per hour or defects per batch. Example application: In a hypothetical Walmart project involving retail commerce, inventory, and supply-chain operations, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A symmetric bell-shaped distribution defined by mean and standard deviation, where about 68%, 95% and 99.7% of values lie within 1, 2 and 3 standard deviations.

Modelling the count of events in a fixed interval when events occur independently at a constant average rate, such as calls per hour or defects per batch.

A symmetric bell-shaped distribution defined by mean and standard deviation, where about 68%, 95% and 99.7% of values lie within 1, 2 and 3 standard deviations.

A symmetric bell-shaped distribution defined by mean and standard deviation, where about 68%, 95% and 99.7% of values lie within 1, 2 and 3 standard deviations.

Modelling the count of events in a fixed interval when events occur independently at a constant average rate, such as calls per hour or defects per batch.

Frequentists treat parameters as fixed and use long-run frequencies; Bayesians treat parameters as distributions updated with data via priors and posteriors, giving direct probability statements.

Frequentists treat parameters as fixed and use long-run frequencies; Bayesians treat parameters as distributions updated with data via priors and posteriors, giving direct probability statements.

The probability of seeing a result at least this extreme if the null hypothesis were true. A small p-value (for example < 0.05) is evidence against the null hypothesis. Example application: In a hypothetical TCS project involving enterprise IT services, integration, and client delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The probability of seeing a result at least this extreme if the null hypothesis were true. A small p-value (for example < 0.05) is evidence against the null hypothesis. Example application: In a hypothetical Amazon project involving an online marketplace, fulfilment operations, and cloud services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The probability of seeing a result at least this extreme if the null hypothesis were true. A small p-value (for example < 0.05) is evidence against the null hypothesis. Example application: In a hypothetical D. E. Shaw project involving quantitative research and financial technology, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The probability of seeing a result at least this extreme if the null hypothesis were true. A small p-value (for example < 0.05) is evidence against the null hypothesis. Example application: In a hypothetical Paytm project involving payments, merchant services, and financial products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The distribution of sample means approaches a normal distribution as the sample size grows (roughly 30+), whatever the population's shape. It lets us build confidence intervals and run hypothesis tests on means. Example application: In a hypothetical TCS project involving enterprise IT services, integration, and client delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The distribution of sample means approaches a normal distribution as the sample size grows (roughly 30+), whatever the population's shape. It lets us build confidence intervals and run hypothesis tests on means. Example application: In a hypothetical Atlassian project involving team collaboration and developer workflow products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The distribution of sample means approaches a normal distribution as the sample size grows (roughly 30+), whatever the population's shape. It lets us build confidence intervals and run hypothesis tests on means. Example application: In a hypothetical Zepto project involving quick commerce, inventory, and last-mile delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The probability of seeing a result at least as extreme as the observed one if the null hypothesis were true. A small p-value is evidence against the null, not proof that the alternative is true. Example application: In a hypothetical Intuit project involving financial software, tax preparation, and small-business tools, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The probability of seeing a result at least as extreme as the observed one if the null hypothesis were true. A small p-value is evidence against the null, not proof that the alternative is true. Example application: In a hypothetical Visa project involving card networks, authorization, and payment-risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The probability of seeing a result at least as extreme as the observed one if the null hypothesis were true. A small p-value is evidence against the null, not proof that the alternative is true. Example application: In a hypothetical Mastercard project involving card payment networks and fraud prevention, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Define the hypothesis and one primary metric, calculate sample size from the baseline, minimum detectable effect, α and power, randomise users, run for full business cycles, check for sample-ratio mismatch, then test both statistical and practical significance. Example application: In a hypothetical Deloitte project involving consulting, risk advisory, and data-driven client services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Define the hypothesis and one primary metric, calculate sample size from the baseline, minimum detectable effect, α and power, randomise users, run for full business cycles, check for sample-ratio mismatch, then test both statistical and practical significance. Example application: In a hypothetical Bloomberg project involving financial terminals, market data, and news systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Define the hypothesis and one primary metric, calculate sample size from the baseline, minimum detectable effect, α and power, randomise users, run for full business cycles, check for sample-ratio mismatch, then test both statistical and practical significance. Example application: In a hypothetical PhonePe project involving UPI payments, merchant transactions, and financial services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Define the hypothesis and one primary metric, calculate sample size from the baseline, minimum detectable effect, α and power, randomise users, run for full business cycles, check for sample-ratio mismatch, then test both statistical and practical significance. Example application: In a hypothetical CRED project involving member rewards, credit-card payments, and financial products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Define the hypothesis and one primary metric, calculate sample size from the baseline, minimum detectable effect, α and power, randomise users, run for full business cycles, check for sample-ratio mismatch, then test both statistical and practical significance. Example application: In a hypothetical JP Morgan Chase project involving banking, payments, and financial risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A symmetric bell-shaped distribution defined by mean and standard deviation, where about 68%, 95% and 99.7% of values lie within 1, 2 and 3 standard deviations. Example application: In a hypothetical Infosys project involving enterprise consulting, digital platforms, and client transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A symmetric bell-shaped distribution defined by mean and standard deviation, where about 68%, 95% and 99.7% of values lie within 1, 2 and 3 standard deviations. Example application: In a hypothetical Cognizant project involving enterprise applications and healthcare or financial client systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Modelling the count of events in a fixed interval when events occur independently at a constant average rate, such as calls per hour or defects per batch. Example application: In a hypothetical Amazon project involving an online marketplace, fulfilment operations, and cloud services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Covariance shows the direction of a relationship but depends on units. Correlation is covariance scaled to −1…+1, so it shows direction and strength and is unit-free. Example application: In a hypothetical Stripe project involving online payments, billing, and financial infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Covariance shows the direction of a relationship but depends on units. Correlation is covariance scaled to −1…+1, so it shows direction and strength and is unit-free. Example application: In a hypothetical PhonePe project involving UPI payments, merchant transactions, and financial services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Pearson measures linear relationships on continuous data; Spearman uses ranks, so it captures monotonic relationships and is less sensitive to outliers. Example application: In a hypothetical PhonePe project involving UPI payments, merchant transactions, and financial services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The square root of the variance; it measures how far values typically sit from the mean, in the same units as the data. Example application: In a hypothetical Meesho project involving social commerce, seller listings, and logistics, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A symmetric bell-shaped distribution defined by mean and standard deviation, where about 68%, 95% and 99.7% of values lie within 1, 2 and 3 standard deviations. Example application: In a hypothetical Paytm project involving payments, merchant services, and financial products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

P(A given B) = P(B given A) x P(A) / P(B). In a medical test it converts test accuracy and disease prevalence into the real chance a positive result means disease. Example application: In a hypothetical Accenture project involving multi-client consulting, cloud transformation, and process automation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

P(A given B) = P(B given A) x P(A) / P(B). In a medical test it converts test accuracy and disease prevalence into the real chance a positive result means disease. Example application: In a hypothetical Capgemini project involving client delivery, data platforms, and enterprise transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It checks whether observed category counts differ from expected counts, used for independence between two categorical variables and for goodness of fit. Example application: In a hypothetical Salesforce project involving CRM, sales operations, and customer data workflows, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

P(A given B) = P(B given A) x P(A) / P(B). In a medical test it converts test accuracy and disease prevalence into the real chance a positive result means disease. Example application: In a hypothetical Tech Mahindra project involving telecom platforms, IT services, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A range built from sample data that would contain the true parameter in a stated percentage (for example 95%) of repeated samples. It is not the probability that the parameter lies inside this one interval.

Define the hypothesis and one primary metric, calculate sample size from the baseline, minimum detectable effect, α and power, randomise users, run for full business cycles, check for sample-ratio mismatch, then test both statistical and practical significance. Example application: In a hypothetical Visa project involving card networks, authorization, and payment-risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Frequentists treat parameters as fixed and use long-run frequencies; Bayesians treat parameters as distributions updated with data via priors and posteriors, giving direct probability statements.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only. Example application: In a hypothetical Cognizant project involving enterprise applications and healthcare or financial client systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only. Example application: In a hypothetical Capgemini project involving client delivery, data platforms, and enterprise transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only. Example application: In a hypothetical PhonePe project involving UPI payments, merchant transactions, and financial services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only. Example application: In a hypothetical CRED project involving member rewards, credit-card payments, and financial products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only. Example application: In a hypothetical Intuit project involving financial software, tax preparation, and small-business tools, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only. Example application: In a hypothetical JP Morgan Chase project involving banking, payments, and financial risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only. Example application: In a hypothetical HSBC project involving international banking, risk, and cross-border payments, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only. Example application: In a hypothetical Tech Mahindra project involving telecom platforms, IT services, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation. Example application: In a hypothetical TCS project involving enterprise IT services, integration, and client delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation. Example application: In a hypothetical Infosys project involving enterprise consulting, digital platforms, and client transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation. Example application: In a hypothetical Cognizant project involving enterprise applications and healthcare or financial client systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation. Example application: In a hypothetical Amazon project involving an online marketplace, fulfilment operations, and cloud services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation. Example application: In a hypothetical PhonePe project involving UPI payments, merchant transactions, and financial services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation. Example application: In a hypothetical Razorpay project involving payment gateway, settlement, and merchant integrations, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation. Example application: In a hypothetical Intuit project involving financial software, tax preparation, and small-business tools, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation. Example application: In a hypothetical Cisco project involving networking, security, and collaboration infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation.

Predictors that are highly correlated make coefficients unstable. Detect with VIF above about 5-10, then drop or combine features, use PCA or apply regularisation.

Pearson measures linear relationships on continuous data; Spearman uses ranks, so it captures monotonic relationships and is less sensitive to outliers. Example application: In a hypothetical Adobe project involving creative software, document workflows, and subscriptions, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only. Example application: In a hypothetical Zomato project involving food ordering, restaurant discovery, and delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

They build a list in one readable expression, for example [x*x for x in data if x > 0]. They are usually faster than an equivalent for-loop with append. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only.

Bias occurs when the sample does not represent the population. Reduce it with random or stratified sampling, higher response rates and by checking sample demographics against the population. Example application: In a hypothetical Wipro project involving managed IT services, cloud migration, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A shallow copy duplicates the outer object but shares nested objects. A deep copy (copy.deepcopy) recursively duplicates everything, so changes to nested items do not leak. Example application: In a hypothetical Cognizant project involving enterprise applications and healthcare or financial client systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A decorator wraps a function to add behaviour such as logging, timing or caching without changing its code. It is applied with the @name syntax above the function. Example application: In a hypothetical Apple project involving consumer devices, App Store workflows, and subscription services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A shallow copy duplicates the outer object but shares nested objects. A deep copy (copy.deepcopy) recursively duplicates everything, so changes to nested items do not leak. Example application: In a hypothetical Bloomberg project involving financial terminals, market data, and news systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

df.groupby('region').agg(total=('sales','sum'), avg=('sales','mean')) gives named aggregations per group in one pass. Example application: In a hypothetical PayPal project involving digital payments, merchant checkout, and transaction risk, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

merge joins on columns, SQL-style (how='inner', 'left', ...). join joins on the index. concat stacks DataFrames along rows or columns. Example application: In a hypothetical Flipkart project involving e-commerce marketplace, seller tools, and fulfilment, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

pd.pivot_table(df, index='region', columns='month', values='sales', aggfunc='sum') reshapes long data into a summary matrix; melt() does the reverse. Example application: In a hypothetical Razorpay project involving payment gateway, settlement, and merchant integrations, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Use the IQR rule (below Q1-1.5 IQR or above Q3+1.5 IQR), z-scores beyond 3, box plots or model-based methods such as Isolation Forest, then decide whether to cap, remove or keep them. Example application: In a hypothetical JP Morgan Chase project involving banking, payments, and financial risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A range built from sample data that would contain the true parameter in a stated percentage (for example 95%) of repeated samples. It is not the probability that the parameter lies inside this one interval. Example application: In a hypothetical Visa project involving card networks, authorization, and payment-risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

*args collects extra positional arguments into a tuple; **kwargs collects extra keyword arguments into a dict. They make function signatures flexible and are common in wrappers and decorators. Example application: In a hypothetical Meta project involving social platforms, messaging, and advertising systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

== compares values for equality, while is checks whether two names point to the same object in memory. Use is mainly for None checks. Example application: In a hypothetical Flipkart project involving e-commerce marketplace, seller tools, and fulfilment, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A method is a function that belongs to an object or class and receives the instance (self) as its first argument. Example application: In a hypothetical Flipkart project involving e-commerce marketplace, seller tools, and fulfilment, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A measure of asymmetry. Right-skewed data has a long tail on the right (mean greater than median); left-skewed is the opposite. Log transforms often reduce right skew. Example application: In a hypothetical Goldman Sachs project involving investment banking, markets, and risk analytics, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A measure of asymmetry. Right-skewed data has a long tail on the right (mean greater than median); left-skewed is the opposite. Log transforms often reduce right skew. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A measure of asymmetry. Right-skewed data has a long tail on the right (mean greater than median); left-skewed is the opposite. Log transforms often reduce right skew. Example application: In a hypothetical HSBC project involving international banking, risk, and cross-border payments, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

loc selects by label (row and column names, inclusive of the end label); iloc selects by integer position (end-exclusive).

Correlation measures linear association between variables (-1 to 1). A third factor, reverse causality or chance can create it, so causation needs controlled experiments. Example application: In a hypothetical Salesforce project involving CRM, sales operations, and customer data workflows, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It checks whether observed category counts differ from expected counts, used for independence between two categorical variables and for goodness of fit. Example application: In a hypothetical Swiggy project involving food delivery, logistics, and quick commerce, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Small anonymous one-line functions, handy as arguments to sort, map, filter or Pandas apply. Example application: In a hypothetical Paytm project involving payments, merchant services, and financial products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Measure first with cProfile or timeit, then replace loops with NumPy or Pandas vectorised calls, use built-ins, caching and better data structures.

A mutex in CPython that lets only one thread execute Python bytecode at a time. Use multiprocessing for CPU-bound work and threads or asyncio for I/O-bound work.

Define the hypothesis and one primary metric, calculate sample size from the baseline, minimum detectable effect, α and power, randomise users, run for full business cycles, check for sample-ratio mismatch, then test both statistical and practical significance. Example application: In a hypothetical TCS project involving enterprise IT services, integration, and client delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam. Example application: In a hypothetical Paytm project involving payments, merchant services, and financial products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam.

It fits a line minimising squared error. Assumes linearity, independent errors, constant variance (homoscedasticity), normally distributed residuals and no strong multicollinearity. Example application: In a hypothetical TCS project involving enterprise IT services, integration, and client delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It fits a line minimising squared error. Assumes linearity, independent errors, constant variance (homoscedasticity), normally distributed residuals and no strong multicollinearity. Example application: In a hypothetical Wipro project involving managed IT services, cloud migration, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It models the log-odds of a binary outcome as a linear function and passes it through a sigmoid to give probabilities, trained by maximising likelihood. Example application: In a hypothetical Meta project involving social platforms, messaging, and advertising systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It models the log-odds of a binary outcome as a linear function and passes it through a sigmoid to give probabilities, trained by maximising likelihood. Example application: In a hypothetical LinkedIn project involving professional networking, job discovery, and recruiting products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It fits a line minimising squared error. Assumes linearity, independent errors, constant variance (homoscedasticity), normally distributed residuals and no strong multicollinearity. Example application: In a hypothetical Uber project involving ride-hailing, driver operations, and real-time dispatch, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It fits a line minimising squared error. Assumes linearity, independent errors, constant variance (homoscedasticity), normally distributed residuals and no strong multicollinearity. Example application: In a hypothetical Flipkart project involving e-commerce marketplace, seller tools, and fulfilment, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It models the log-odds of a binary outcome as a linear function and passes it through a sigmoid to give probabilities, trained by maximising likelihood. Example application: In a hypothetical Flipkart project involving e-commerce marketplace, seller tools, and fulfilment, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It fits a line minimising squared error. Assumes linearity, independent errors, constant variance (homoscedasticity), normally distributed residuals and no strong multicollinearity. Example application: In a hypothetical ServiceNow project involving IT service management and enterprise workflow automation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It fits a line minimising squared error. Assumes linearity, independent errors, constant variance (homoscedasticity), normally distributed residuals and no strong multicollinearity. Example application: In a hypothetical Barclays project involving retail banking, trading, and financial compliance, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It models the log-odds of a binary outcome as a linear function and passes it through a sigmoid to give probabilities, trained by maximising likelihood. Example application: In a hypothetical HCLTech project involving enterprise technology services and infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It models the log-odds of a binary outcome as a linear function and passes it through a sigmoid to give probabilities, trained by maximising likelihood.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam. Example application: In a hypothetical Capgemini project involving client delivery, data platforms, and enterprise transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam. Example application: In a hypothetical Microsoft project involving developer tools, cloud services, and productivity applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam. Example application: In a hypothetical Uber project involving ride-hailing, driver operations, and real-time dispatch, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam. Example application: In a hypothetical Stripe project involving online payments, billing, and financial infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam. Example application: In a hypothetical Salesforce project involving CRM, sales operations, and customer data workflows, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam. Example application: In a hypothetical Meesho project involving social commerce, seller listings, and logistics, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam. Example application: In a hypothetical Swiggy project involving food delivery, logistics, and quick commerce, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam. Example application: In a hypothetical Goldman Sachs project involving investment banking, markets, and risk analytics, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam. Example application: In a hypothetical Mastercard project involving card payment networks and fraud prevention, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam. Example application: In a hypothetical HCLTech project involving enterprise technology services and infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam. Example application: In a hypothetical Tech Mahindra project involving telecom platforms, IT services, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data. Example application: In a hypothetical Amazon project involving an online marketplace, fulfilment operations, and cloud services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data. Example application: In a hypothetical Apple project involving consumer devices, App Store workflows, and subscription services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data. Example application: In a hypothetical PayPal project involving digital payments, merchant checkout, and transaction risk, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data. Example application: In a hypothetical Salesforce project involving CRM, sales operations, and customer data workflows, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data. Example application: In a hypothetical Adobe project involving creative software, document workflows, and subscriptions, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data. Example application: In a hypothetical Zepto project involving quick commerce, inventory, and last-mile delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data. Example application: In a hypothetical Walmart project involving retail commerce, inventory, and supply-chain operations, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data. Example application: In a hypothetical Visa project involving card networks, authorization, and payment-risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data. Example application: In a hypothetical Barclays project involving retail banking, trading, and financial compliance, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data. Example application: In a hypothetical Tech Mahindra project involving telecom platforms, IT services, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical Cognizant project involving enterprise applications and healthcare or financial client systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical Capgemini project involving client delivery, data platforms, and enterprise transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical Amazon project involving an online marketplace, fulfilment operations, and cloud services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical Meta project involving social platforms, messaging, and advertising systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical Microsoft project involving developer tools, cloud services, and productivity applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical Uber project involving ride-hailing, driver operations, and real-time dispatch, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical Stripe project involving online payments, billing, and financial infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical Flipkart project involving e-commerce marketplace, seller tools, and fulfilment, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical Zepto project involving quick commerce, inventory, and last-mile delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical Intuit project involving financial software, tax preparation, and small-business tools, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical JP Morgan Chase project involving banking, payments, and financial risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical Visa project involving card networks, authorization, and payment-risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context. Example application: In a hypothetical HSBC project involving international banking, risk, and cross-border payments, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context.

Putting features on comparable ranges using standardisation or min-max scaling. Distance-based and gradient-based models (KNN, SVM, regression, neural nets) need it; tree models do not. Example application: In a hypothetical Infosys project involving enterprise consulting, digital platforms, and client transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Define churn and the prediction window, build behavioural features, split by time, train and compare models, evaluate with recall and lift, explain drivers and monitor drift after deployment. Example application: In a hypothetical Amazon project involving an online marketplace, fulfilment operations, and cloud services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Define churn and the prediction window, build behavioural features, split by time, train and compare models, evaluate with recall and lift, explain drivers and monitor drift after deployment. Example application: In a hypothetical Atlassian project involving team collaboration and developer workflow products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Define churn and the prediction window, build behavioural features, split by time, train and compare models, evaluate with recall and lift, explain drivers and monitor drift after deployment. Example application: In a hypothetical Stripe project involving online payments, billing, and financial infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Discriminative models learn P(y given x) directly (logistic regression, SVM). Generative models learn the joint P(x, y) and can generate data (Naive Bayes, GANs). Example application: In a hypothetical PayPal project involving digital payments, merchant checkout, and transaction risk, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Tie metrics to business outcomes such as cost saved, show feature importance or SHAP values, give example predictions and be transparent about limitations. Example application: In a hypothetical CRED project involving member rewards, credit-card payments, and financial products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Tie metrics to business outcomes such as cost saved, show feature importance or SHAP values, give example predictions and be transparent about limitations. Example application: In a hypothetical Zepto project involving quick commerce, inventory, and last-mile delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Tie metrics to business outcomes such as cost saved, show feature importance or SHAP values, give example predictions and be transparent about limitations. Example application: In a hypothetical ServiceNow project involving IT service management and enterprise workflow automation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Tie metrics to business outcomes such as cost saved, show feature importance or SHAP values, give example predictions and be transparent about limitations. Example application: In a hypothetical Morgan Stanley project involving wealth management, trading, and investment platforms, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Package the model behind an API with versioning, track latency and errors, monitor input drift and prediction quality, and automate retraining with rollback. Example application: In a hypothetical Cisco project involving networking, security, and collaboration infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Package the model behind an API with versioning, track latency and errors, monitor input drift and prediction quality, and automate retraining with rollback. Example application: In a hypothetical Visa project involving card networks, authorization, and payment-risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Tie metrics to business outcomes such as cost saved, show feature importance or SHAP values, give example predictions and be transparent about limitations. Example application: In a hypothetical HSBC project involving international banking, risk, and cross-border payments, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Package the model behind an API with versioning, track latency and errors, monitor input drift and prediction quality, and automate retraining with rollback. Example application: In a hypothetical HCLTech project involving enterprise technology services and infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Tie metrics to business outcomes such as cost saved, show feature importance or SHAP values, give example predictions and be transparent about limitations. Example application: In a hypothetical Tech Mahindra project involving telecom platforms, IT services, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Discriminative models learn P(y given x) directly (logistic regression, SVM). Generative models learn the joint P(x, y) and can generate data (Naive Bayes, GANs).

Layers of connected neurons that apply weights, biases and activation functions to learn complex patterns, trained by backpropagation. Example application: In a hypothetical Infosys project involving enterprise consulting, digital platforms, and client transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Layers of connected neurons that apply weights, biases and activation functions to learn complex patterns, trained by backpropagation. Example application: In a hypothetical Deloitte project involving consulting, risk advisory, and data-driven client services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Layers of connected neurons that apply weights, biases and activation functions to learn complex patterns, trained by backpropagation. Example application: In a hypothetical Microsoft project involving developer tools, cloud services, and productivity applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Layers of connected neurons that apply weights, biases and activation functions to learn complex patterns, trained by backpropagation. Example application: In a hypothetical LinkedIn project involving professional networking, job discovery, and recruiting products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Layers of connected neurons that apply weights, biases and activation functions to learn complex patterns, trained by backpropagation. Example application: In a hypothetical Flipkart project involving e-commerce marketplace, seller tools, and fulfilment, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Layers of connected neurons that apply weights, biases and activation functions to learn complex patterns, trained by backpropagation. Example application: In a hypothetical JP Morgan Chase project involving banking, payments, and financial risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Layers of connected neurons that apply weights, biases and activation functions to learn complex patterns, trained by backpropagation. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Layers of connected neurons that apply weights, biases and activation functions to learn complex patterns, trained by backpropagation.

Layers of connected neurons that apply weights, biases and activation functions to learn complex patterns, trained by backpropagation.

Layers of connected neurons that apply weights, biases and activation functions to learn complex patterns, trained by backpropagation.

Start with a sensible default such as 1e-3 for Adam, run a learning-rate range test, and use schedulers or warm-up while watching the loss curve.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs. Example application: In a hypothetical Deloitte project involving consulting, risk advisory, and data-driven client services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs. Example application: In a hypothetical Atlassian project involving team collaboration and developer workflow products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs. Example application: In a hypothetical Tower Research Capital project involving low-latency trading and market data systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs. Example application: In a hypothetical PayPal project involving digital payments, merchant checkout, and transaction risk, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs. Example application: In a hypothetical Zomato project involving food ordering, restaurant discovery, and delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs. Example application: In a hypothetical Swiggy project involving food delivery, logistics, and quick commerce, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs. Example application: In a hypothetical ServiceNow project involving IT service management and enterprise workflow automation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs. Example application: In a hypothetical Morgan Stanley project involving wealth management, trading, and investment platforms, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs. Example application: In a hypothetical Cognizant project involving enterprise applications and healthcare or financial client systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs. Example application: In a hypothetical Amazon project involving an online marketplace, fulfilment operations, and cloud services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs. Example application: In a hypothetical Microsoft project involving developer tools, cloud services, and productivity applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs. Example application: In a hypothetical Stripe project involving online payments, billing, and financial infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs. Example application: In a hypothetical PayPal project involving digital payments, merchant checkout, and transaction risk, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs. Example application: In a hypothetical ServiceNow project involving IT service management and enterprise workflow automation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs. Example application: In a hypothetical Walmart project involving retail commerce, inventory, and supply-chain operations, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs. Example application: In a hypothetical HCLTech project involving enterprise technology services and infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs. Example application: In a hypothetical Tech Mahindra project involving telecom platforms, IT services, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs.

Splitting text into units such as words, subwords or characters that are mapped to ids; subword methods like BPE handle rare words well. Example application: In a hypothetical Google project involving search, advertising, maps, and cloud products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Splitting text into units such as words, subwords or characters that are mapped to ids; subword methods like BPE handle rare words well. Example application: In a hypothetical Visa project involving card networks, authorization, and payment-risk systems, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Term Frequency × Inverse Document Frequency: a word scores high if it is frequent in one document but rare across all documents. A classic way to turn text into numeric features. Example application: In a hypothetical Salesforce project involving CRM, sales operations, and customer data workflows, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It applies Bayes' theorem assuming features are independent given the class. The assumption is rarely true but the model is fast and works well for text. Example application: In a hypothetical HSBC project involving international banking, risk, and cross-border payments, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It applies Bayes' theorem assuming features are independent given the class. The assumption is rarely true but the model is fast and works well for text.

It applies Bayes' theorem assuming features are independent given the class. The assumption is rarely true but the model is fast and works well for text.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs. Example application: In a hypothetical Infosys project involving enterprise consulting, digital platforms, and client transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs. Example application: In a hypothetical Accenture project involving multi-client consulting, cloud transformation, and process automation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs. Example application: In a hypothetical Deloitte project involving consulting, risk advisory, and data-driven client services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs. Example application: In a hypothetical Apple project involving consumer devices, App Store workflows, and subscription services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs. Example application: In a hypothetical Barclays project involving retail banking, trading, and financial compliance, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

They build a list in one readable expression, for example [x*x for x in data if x > 0]. They are usually faster than an equivalent for-loop with append. Example application: In a hypothetical Wipro project involving managed IT services, cloud migration, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Information from outside the training window, such as future data or test statistics, enters training and inflates performance. Split first, then fit scalers and encoders on training data only. Example application: In a hypothetical Morgan Stanley project involving wealth management, trading, and investment platforms, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

List: ordered and mutable. Tuple: ordered and immutable (hashable, slightly faster). Set: unordered, unique items, fast membership checks. Dict: key–value pairs with fast lookup by key.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only.

An iterable can produce an iterator (it has __iter__); an iterator has __next__ and remembers its position. A for-loop calls iter() on the iterable and then next() repeatedly. Example application: In a hypothetical Google project involving search, advertising, maps, and cloud products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It computes the gradient of the loss with respect to every weight by applying the chain rule backwards through the network, so gradient descent can update them. Example application: In a hypothetical Salesforce project involving CRM, sales operations, and customer data workflows, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

SVM finds the maximum-margin separating hyperplane. Kernels such as RBF compute similarity in a higher-dimensional space implicitly, so non-linear boundaries are possible. Example application: In a hypothetical Razorpay project involving payment gateway, settlement, and merchant integrations, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Term Frequency × Inverse Document Frequency: a word scores high if it is frequent in one document but rare across all documents. A classic way to turn text into numeric features. Example application: In a hypothetical Intuit project involving financial software, tax preparation, and small-business tools, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

(df.assign(rev=df.price * df.qty) .groupby(['region', 'product'])['rev'].sum() .reset_index() .sort_values('rev', ascending=False) .groupby('region').head(3)) Example application: In a hypothetical Walmart project involving retail commerce, inventory, and supply-chain operations, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A shallow copy duplicates the outer object but shares nested objects. A deep copy (copy.deepcopy) recursively duplicates everything, so changes to nested items do not leak. Example application: In a hypothetical Oracle project involving enterprise databases, cloud infrastructure, and business applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Use stratified splits, class weights, resampling (SMOTE or undersampling) and threshold tuning, and judge the model with PR-AUC or F1 instead of accuracy.

An optimisation method that updates weights in the opposite direction of the loss gradient. The learning rate sets the step size: too high and it diverges, too low and it is slow. Variants include SGD, mini-batch and Adam.

It normalises layer inputs per mini-batch and then rescales them, stabilising and speeding up training and allowing higher learning rates.

Regularization adds a penalty on weights to reduce overfitting. L1 (Lasso) adds |w| and can push features to exactly zero. L2 (Ridge) adds w² and shrinks weights smoothly.

Define churn and the prediction window → build usage and billing features → handle imbalance → train gradient boosting → evaluate with PR-AUC and lift → explain with SHAP → deploy as batch scoring or an API → monitor drift → hand top-risk users to the retention team. Example application: In a hypothetical HCLTech project involving enterprise technology services and infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

int, float, complex, bool, str, list, tuple, set, dict and NoneType. Mutable ones are list, set and dict; the others are immutable. Example application: In a hypothetical Infosys project involving enterprise consulting, digital platforms, and client transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The rule that lets arrays of different shapes operate together by virtually stretching the smaller one, for example adding a scalar or a (3,) vector to a (4,3) matrix.

Term Frequency × Inverse Document Frequency: a word scores high if it is frequent in one document but rare across all documents. A classic way to turn text into numeric features. Example application: In a hypothetical Amazon project involving an online marketplace, fulfilment operations, and cloud services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

ReLU for hidden layers, sigmoid for binary outputs, softmax for multi-class outputs and tanh when zero-centred outputs help; Leaky ReLU avoids dead neurons. Example application: In a hypothetical Microsoft project involving developer tools, cloud services, and productivity applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

An optimisation method that updates weights in the opposite direction of the loss gradient. The learning rate sets the step size: too high and it diverges, too low and it is slow. Variants include SGD, mini-batch and Adam. Example application: In a hypothetical Apple project involving consumer devices, App Store workflows, and subscription services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The ROC curve plots true positive rate against false positive rate across thresholds. AUC is the probability that a random positive is ranked above a random negative; 0.5 is chance. Example application: In a hypothetical Uber project involving ride-hailing, driver operations, and real-time dispatch, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

The ROC curve plots true positive rate against false positive rate across thresholds. AUC is the probability that a random positive is ranked above a random negative; 0.5 is chance. Example application: In a hypothetical D. E. Shaw project involving quantitative research and financial technology, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Activations add non-linearity so networks can learn complex patterns. ReLU (max(0, x)) is cheap and reduces vanishing gradients compared with sigmoid/tanh; Leaky ReLU avoids 'dead' neurons. Example application: In a hypothetical Stripe project involving online payments, billing, and financial infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It tries candidate splits and picks the one that most reduces impurity — Gini or entropy (information gain) for classification, variance / MSE for regression. Example application: In a hypothetical PayPal project involving digital payments, merchant checkout, and transaction risk, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Term Frequency × Inverse Document Frequency: a word scores high if it is frequent in one document but rare across all documents. A classic way to turn text into numeric features. Example application: In a hypothetical Razorpay project involving payment gateway, settlement, and merchant integrations, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It tries candidate splits and picks the one that most reduces impurity — Gini or entropy (information gain) for classification, variance / MSE for regression. Example application: In a hypothetical Zepto project involving quick commerce, inventory, and last-mile delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A decorator wraps a function to add behaviour such as logging, timing or caching without changing its code. It is applied with the @name syntax above the function. Example application: In a hypothetical Morgan Stanley project involving wealth management, trading, and investment platforms, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Use filter methods (correlation, chi-square), wrapper methods (RFE), embedded methods (Lasso, tree importance) and domain knowledge, validating each choice with cross-validation. Example application: In a hypothetical HSBC project involving international banking, risk, and cross-border payments, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

An iterable can produce an iterator (it has __iter__); an iterator has __next__ and remembers its position. A for-loop calls iter() on the iterable and then next() repeatedly.

Define churn and the prediction window, build behavioural features, split by time, train and compare models, evaluate with recall and lift, explain drivers and monitor drift after deployment. Example application: In a hypothetical Cisco project involving networking, security, and collaboration infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A table of true positives, false positives, true negatives and false negatives from which accuracy, precision, recall and F1 are calculated. Example application: In a hypothetical Microsoft project involving developer tools, cloud services, and productivity applications, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

df.drop_duplicates(subset=[...], keep='first') removes repeated rows; use duplicated() first to inspect what will be dropped. Example application: In a hypothetical Flipkart project involving e-commerce marketplace, seller tools, and fulfilment, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Layers of connected neurons that apply weights, biases and activation functions to learn complex patterns, trained by backpropagation. Example application: In a hypothetical Razorpay project involving payment gateway, settlement, and merchant integrations, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Splitting text into units such as words, subwords or characters that are mapped to ids; subword methods like BPE handle rare words well. Example application: In a hypothetical Zepto project involving quick commerce, inventory, and last-mile delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A shallow copy duplicates the outer object but shares nested objects. A deep copy (copy.deepcopy) recursively duplicates everything, so changes to nested items do not leak. Example application: In a hypothetical TCS project involving enterprise IT services, integration, and client delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A shallow copy duplicates the outer object but shares nested objects. A deep copy (copy.deepcopy) recursively duplicates everything, so changes to nested items do not leak. Example application: In a hypothetical LinkedIn project involving professional networking, job discovery, and recruiting products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Reusing a model pretrained on a large dataset and fine-tuning it on a smaller task-specific dataset, saving data, time and compute. Example application: In a hypothetical Meesho project involving social commerce, seller listings, and logistics, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It models the log-odds of a binary outcome as a linear function and passes it through a sigmoid to give probabilities, trained by maximising likelihood. Example application: In a hypothetical Zomato project involving food ordering, restaurant discovery, and delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

ReLU for hidden layers, sigmoid for binary outputs, softmax for multi-class outputs and tanh when zero-centred outputs help; Leaky ReLU avoids dead neurons. Example application: In a hypothetical Paytm project involving payments, merchant services, and financial products, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Regularization adds a penalty on weights to reduce overfitting. L1 (Lasso) adds |w| and can push features to exactly zero. L2 (Ridge) adds w² and shrinks weights smoothly. Example application: In a hypothetical ServiceNow project involving IT service management and enterprise workflow automation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Use filter methods (correlation, chi-square), wrapper methods (RFE), embedded methods (Lasso, tree importance) and domain knowledge, validating each choice with cross-validation.

merge joins on columns, SQL-style (how='inner', 'left', ...). join joins on the index. concat stacks DataFrames along rows or columns.

The ROC curve plots true positive rate against false positive rate across thresholds. AUC is the probability that a random positive is ranked above a random negative; 0.5 is chance.

Package the model behind an API with versioning, track latency and errors, monitor input drift and prediction quality, and automate retraining with rollback. Example application: In a hypothetical D. E. Shaw project involving quantitative research and financial technology, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Retrieval-Augmented Generation fetches relevant documents and gives them to an LLM as context, reducing hallucination and keeping answers grounded in current private data. Example application: In a hypothetical Tech Mahindra project involving telecom platforms, IT services, and digital transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Train fits the model, validation tunes hyperparameters and model choice, and the untouched test set gives a final unbiased estimate of performance. Example application: In a hypothetical Capgemini project involving client delivery, data platforms, and enterprise transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Information from outside the training window, such as future data or test statistics, enters training and inflates performance. Split first, then fit scalers and encoders on training data only. Example application: In a hypothetical Apple project involving consumer devices, App Store workflows, and subscription services, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Splitting text into units such as words, subwords or characters that are mapped to ids; subword methods like BPE handle rare words well.

The ROC curve plots true positive rate against false positive rate across thresholds. AUC is the probability that a random positive is ranked above a random negative; 0.5 is chance. Example application: In a hypothetical TCS project involving enterprise IT services, integration, and client delivery, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It projects data onto orthogonal directions of maximum variance to reduce dimensions, remove correlation and speed up models, at some cost of interpretability. Example application: In a hypothetical Uber project involving ride-hailing, driver operations, and real-time dispatch, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

It computes the gradient of the loss with respect to every weight by applying the chain rule backwards through the network, so gradient descent can update them. Example application: In a hypothetical Stripe project involving online payments, billing, and financial infrastructure, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

A forest averages many trees trained on bootstrapped rows and random feature subsets, which cuts variance and overfitting at the cost of interpretability. Example application: In a hypothetical Flipkart project involving e-commerce marketplace, seller tools, and fulfilment, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

(df.assign(rev=df.price * df.qty) .groupby(['region', 'product'])['rev'].sum() .reset_index() .sort_values('rev', ascending=False) .groupby('region').head(3))

Define churn and the prediction window, build behavioural features, split by time, train and compare models, evaluate with recall and lift, explain drivers and monitor drift after deployment. Example application: In a hypothetical Capgemini project involving client delivery, data platforms, and enterprise transformation, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Package the model behind an API with versioning, track latency and errors, monitor input drift and prediction quality, and automate retraining with rollback. Example application: In a hypothetical Morgan Stanley project involving wealth management, trading, and investment platforms, compare with a baseline, prevent leakage, evaluate segment performance, check calibration or explainability when relevant, and monitor drift.

Read in chunks or use usecols and dtypes, convert strings to category, avoid apply with row loops in favour of vectorised ops, and consider Polars, Dask or Parquet storage.