Float64
Float64 is a 64-bit floating-point numeric type, commonly represented using the IEEE 754 binary64 format.
Modern C, Third Edition
double is a C floating-point type providing at least the precision and range required for double-precision real-number calculations.
- How it is used
- Uses double for real floating-point values and for decimal literals requiring floating-point semantics.
- How deep it goes
- Defines precision and approximation, then compares literal spellings, but stops short of detailing floating-point storage internals.
- Examples
- Uses 0.2, extended decimal spellings, and hexadecimal floating-point literals to demonstrate approximation and representation.
- What you can do after
- Evaluate/understand why double literals can have rounded effective values and how hexadecimal notation changes their spelling.
DuckDB in Action
A double is a floating-point numeric data type that stores values with approximately twice the precision of a single-precision value.
- How it is used
- Corrects inferred string fields so nested JSON data supports numeric analysis in DuckDB.
- How deep it goes
- Explains schema inspection, whole-STRUCT casting, and verification, but defers floating-point representation internals.
- Examples
- Uses football shot coordinates, expected-goals values, corrected schemas, and team-level xG totals.
- What you can do after
- Operate/configure DuckDB to cast inferred JSON fields and perform numeric queries over flattened records.
100 Java Mistakes and How to Avoid Them
- How it is used
- Uses float64 for numeric algorithms while exposing representation, boundary-constant, and precision-conversion traps.
- How deep it goes
- Explains signed zero, extrema constants, and widening precision loss with code consequences, but does not develop full floating-point arithmetic.
- Examples
- Uses signed-zero multiplication, boxed equality, a flawed maximum search, and long values that collide after double conversion.
- What you can do after
- Diagnose float64 bugs involving signed zero, incorrect extrema initialization, and silently rounded widening conversions.
Read how 100 Java Mistakes and How to Avoid Them explains float64 →
Pandas Workout: 200 exercises to make you a stronger data analyst
float64 is a 64-bit floating-point data type commonly used as the default numeric precision for pandas columns.
- How it is used
- Handles decimal values and missing data in pandas columns during data-frame manipulation and analysis.
- How deep it goes
- Explains dtype coercion, floating-point surprises, and precision tradeoffs, but stops short of floating-point internals.
- Examples
- Uses a product-sales data frame and NYC taxi totals to demonstrate missing values, arithmetic differences, and alternative precision.
- What you can do after
- Diagnose unexpected float64 conversion and evaluate whether higher precision justifies additional memory and reduced portability.
Read how Pandas Workout: 200 exercises to make you a stronger data analyst explains float64 →
ScyllaDB in Action
DOUBLE is a ScyllaDB floating-point data type for storing double-precision numeric values.
- How it is used
- Stores decimal-valued fields when 64-bit floating-point representation fits the application’s precision requirements.
- How deep it goes
- Defines size, encoding, usage, and precision trade-offs, but does not detail floating-point internals or numeric limits.
- Examples
- Demonstrates DOUBLE with the CQL value 567.8 and retrieves it alongside FLOAT value 123.4.
- What you can do after
- Apply it via CQL by inserting and selecting double-precision values, while recognizing when DECIMAL is safer.
100 Go Mistakes and How to Avoid Them
float64 is Go's 64-bit floating-point type with an 11-bit exponent, a 52-bit mantissa, and a sign bit.
- How it is used
- Explains fractional values, approximation effects, arithmetic surprises, and numeric types produced by JSON unmarshaling.
- How deep it goes
- Defines the IEEE-754 bit layout and demonstrates rounding, accumulation order, expression order, special values, and JSON conversion.
- Examples
- Uses multiplication, repeated addition, reassociated expressions, special-value divisions, and decoded JSON identifiers.
- What you can do after
- Diagnose unexpected rounding, accumulation differences, special values, and numeric type assertions in decoded maps.
Read how 100 Go Mistakes and How to Avoid Them explains float64 →
Deep Learning with JAX
A 64-bit floating-point numeric type used as a higher-precision array data type in numerical computing.
- How it is used
- Uses float64 for high-precision scientific calculations when float32's precision is insufficient.
- How deep it goes
- Explains IEEE 754 layout, precision, range, JAX defaults, startup configuration, and backend caveats.
- Examples
- Demonstrates enabling x64 mode, constructing a float64 JAX array, and checking its dtype.
- What you can do after
- Operate/configure JAX to create verified float64 arrays for higher-precision numerical work.
Related concepts
- floating-point arithmetickind of · Floating-point arithmetic is computation using finite-precision representations of real numbers, in which rounding and limited range can make results differ from exact real-number arithmetic.
- IEEE 754implements · IEEE 754 is an international standard specifying floating-point formats, arithmetic operations, rounding rules, and exceptional values.
- FP32alternative to · FP32 is the IEEE 754 binary32 single-precision floating-point format, which represents approximate real numbers using 32 bits.
- data conversionproduced by · Data conversion is the transformation of a value or data from one type, format, or representation to another, explicitly or implicitly.
- float16alternative to · Float16 is a 16-bit floating-point numeric format that represents values with less precision and storage than single-precision formats such as float32.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.