Concept Glossary: I¶

This page lists English glossary entries for this letter. Entry bodies are assembled from term source files.

iloc¶

• Meaning: iloc is the Pandas selector used to choose rows and columns by integer position. If loc selects by label, iloc selects by position, such as the third row or the second column.
• Why it matters: Table code becomes much clearer when readers separate label-based selection from position-based selection. iloc also prevents a common beginner mistake: assuming that a numeric-looking index label is always the same thing as row position.
• Related concepts: loc, index, column
• Core Section: P2-12.2
• Appears in: P2-12.3

image recognition¶

• Meaning: Image recognition is the task of predicting what is present in an input image or which category the image belongs to. It connects arrays of pixels to meaningful visual categories.
• Why it matters: Image recognition is a clear example of the shift from hand-crafted visual cues to learned representations. It helps readers see that image input may be divided into classification, detection, segmentation, and other more specific visual tasks.
• Related concepts: CNN, convolutional neural network, hand-crafted features, learned representation
• Core Section: P1-9.1
• Appears in: P1-9.2, P1-9.3, P1-10.1, P5-2.1, P5-11.1, P5-11.2

improvement plan¶

• Meaning: An improvement plan is an action list that turns problems found in retrospectives or failure records into the next sequence of changes. It specifies what to fix first, what to defer, and what additional checks to add.
• Why it matters: Review notes and incident records do not improve a project by themselves. An improvement plan converts observations into prioritized work so the next iteration starts from concrete actions rather than vague regret.
• Related concepts: retrospective, priority, next action, failure type, incident record
• Core Section: P7-7.2
• Appears in: P7-3.2, P7-summary

in-context learning¶

• Meaning: In-context learning is the behavior or usage pattern in which a model changes its output behavior based on instructions and examples in the prompt, without updating model weights.
• Why it matters: This concept explains why natural-language instructions and few-shot examples matter in LLM use. It also separates changing the prompt from fine-tuning or retraining the model.
• Related concepts: pretraining, fine-tuning, GPT
• Core Section: P1-12.2
• Appears in: P1-11.3

incident record¶

• Meaning: An incident record is an operational note that captures a problem observed during deployment or operation, including date, signals, possible cause, reader impact, and next action.
• Why it matters: The same symptom can have different causes. Incident records keep evidence, interpretation boundaries, and follow-up actions together so later troubleshooting does not start from memory alone.
• Related concepts: failure, failure type, status inspection, next action, retrospective
• Core Section: P7-7.2
• Appears in: P7-index, P7-summary

incomplete information¶

• Meaning: Incomplete information is a situation where the facts needed for judgment or choice are not fully available at the start. The rules or goal may exist, but some necessary input facts are missing.
• Why it matters: When key information is missing, a system may need probability, search, retrieval, or human review rather than just more rules. The concept helps readers distinguish lack of information from lack of logic.
• Related concepts: partial observability, uncertainty, rule-based system
• Core Section: P1-6.1
• Appears in: P1-2.2, P1-7.1

index¶

• Meaning: An index is a position number in an ordered collection, and in Pandas it can also be a label used to identify rows. The same visible value may therefore refer to position or to a row label depending on context.
• Why it matters: Lists, strings, arrays, and tables all rely on index-like selection, but the meaning changes across contexts. Understanding index prevents confusion between choosing the first row and choosing the row labeled 1.
• Related concepts: list, value, loop
• Core Section: P2-8.2
• Appears in: P6-12.2

indexing¶

• Meaning: Indexing is the act of selecting a specific value or location from an array, list, or similar structure. It focuses on picking a particular position, unlike slicing, which keeps a range.
• Why it matters: Indexing and slicing can produce very different shapes and meanings. This distinction helps readers understand whether code selected one value, one row, one column, or a larger substructure.
• Related concepts: slicing, axis, matrix
• Core Section: P2-11.2
• Appears in: P2-11.3, P2-11.4, P2-12.2

indicator¶

• Meaning: An indicator is a number or signal used to observe a direction of change, such as investment, performance, employment, policy, error rate, coverage, or response time.
• Why it matters: Indicators are observations, not reality itself. The concept helps readers separate a signal from the interpretation attached to it and ask what period, baseline, and measurement rule produced the number.
• Related concepts: forecast, scenario, report
• Core Section: P1-17.1
• Appears in: P1-17.2, P1-17.3, P7-1.1, P7-1.3, P7-7.4

inference¶

• Meaning: Inference has multiple meanings. In symbolic AI, it can mean deriving a conclusion from known facts and rules. In machine learning, model inference often means running a trained model on new input to obtain output.
• Why it matters: The same word can refer to reasoning, prediction, or model execution depending on context. Readers need this distinction to avoid mixing a logical conclusion process with the runtime use of a trained model.
• Related concepts: inference engine, training, prediction, generation
• Core Section: P1-5.3
• Appears in: P1-2.1, P1-3.2, P1-5.2, P5-6.1, P5-6.3

inference engine¶

• Meaning: An inference engine is the component that finds and applies rules that match the current facts. It turns stored rules into an execution process that can derive conclusions.
• Why it matters: Rule-based AI is not just a list of rules. The inference engine shows how knowledge representation and rule-application procedure work together to produce a result.
• Related concepts: rule-based system, fact
• Core Section: P1-3.1
• Appears in: P1-2.1

information retrieval¶

• Meaning: Information retrieval is the task of finding documents, passages, or evidence candidates that match a current question or information need. It is often a ranking problem rather than a simple exact string match.
• Why it matters: Retrieval-augmented generation depends on retrieving relevant evidence before generation. Weak retrieval can make the generated answer weak even when the language model is fluent.
• Related concepts: retrieval-augmented generation, RAG, similarity search, search, ranking
• Core Section: P1-13.3
• Appears in: P1-13.4, P1-14.2, P7-5.1, P7-5.2

initialization¶

• Meaning: Initialization is the step of choosing starting values for weights or other parameters before training begins. It sets the numerical starting point from which learning will move.
• Why it matters: Poor starting values can make neurons behave too similarly or make values and gradients unstable. Initialization reminds readers that training success depends not only on data and architecture, but also on the starting numerical layout.
• Related concepts: numerical stability, weight, backpropagation
• Core Section: P5-8.3
• Appears in: P5-3.2

input¶

• Meaning: Input is the information a model actually receives for one computation. It is not the whole outside world or the entire raw dataset, but the values brought into the current run.
• Why it matters: A model can only use the evidence it is given. Input design determines what cues the model can see, what it cannot see, and how preprocessing or feature design affects performance.
• Related concepts: output, data, model, feature, task definition
• Core Section: P1-4.2
• Appears in: P1-4.1

input file¶

• Meaning: An input file is the concrete file read by a practice run or project execution, such as raw data, configuration, or logs. It is the file-level starting point for reproducing the run.
• Why it matters: If the input file is unclear, the same question and code can lead to different interpretations. Recording the input file connects results back to the actual source used in the run.
• Related concepts: input, input unit, run record, reproducibility, working directory
• Core Section: P7-1.1
• Appears in: P7-1.3, P7-2.1, P7-3.1, P7-4.1, P7-5.1, P7-6.1, P7-7.1

input unit¶

• Meaning: An input unit is the rule for deciding what counts as one input case in a project or analysis. It may be one row, a group of rows, a channel-day, a session, or another boundary chosen for the question.
• Why it matters: The input unit changes summaries, baselines, comparisons, and error cases. It helps readers distinguish having raw data from defining the case boundary that the model or analysis will actually use.
• Related concepts: sample, row, summary table, comparison table, task definition
• Core Section: P7-1.1
• Appears in: P7-index, P7-1.3, P7-summary

instruction¶

• Meaning: An instruction is the part of input that tells a model what task to perform. It may ask the model to summarize, classify, compare, translate, critique, or follow another task direction.
• Why it matters: The same context can produce different outputs when the instruction changes. The concept separates providing information from specifying what transformation or action should be performed on that information.
• Related concepts: prompt, context, constraint
• Core Section: P1-12.1
• Appears in: P1-12.2, P1-12.3, P1-14.3, P6-10.1, P6-10.3, P6-summary

instruction tuning¶

• Meaning: Instruction tuning is an additional adaptation stage that helps a pretrained model respond better to natural-language user instructions and expected answer formats.
• Why it matters: A model can know language patterns without automatically following requests well. Instruction tuning helps explain why chat-style LLMs behave more like assistants than plain continuation models.
• Related concepts: fine-tuning, pretraining, prompt
• Core Section: P6-9.1
• Appears in: P6-9.2

interactive mode¶

• Meaning: Interactive mode is an execution style where a user enters commands one at a time and sees results immediately. It is useful for quick checks and short experiments.
• Why it matters: Interactive execution is convenient for exploration, but weaker as a repeatable record than scripts or notebooks. The distinction helps readers choose between quick testing and reproducible execution.
• Related concepts: interpreter, script, prompt
• Core Section: P2-7.3
• Appears in: P2-10.1

intermediate representation¶

• Meaning: An intermediate representation is a human-designed representation placed between raw data and final features so structure, order, or patterns become easier to inspect.
• Why it matters: Some problems are hard to see from raw logs or simple numeric summaries. Intermediate representations let readers compare whether segments, token sequences, or pattern blocks reveal the problem better.
• Related concepts: feature, representation, representation learning
• Core Section: P3-6.2
• Appears in: P3-index, P3-5.1, P3-5.3, P3-6.1, P3-6.3, P3-summary

interpretation¶

• Meaning: Interpretation is a human explanation that attaches meaning to an event or piece of evidence. It goes beyond restating what happened and says how the event should be read.
• Why it matters: Reports and data analysis require separating observed facts from the meaning assigned to them. This reduces exaggeration and helps readers see where different viewpoints may attach different interpretations to the same facts.
• Related concepts: factual claim, column, working hypothesis
• Core Section: P1-17.3
• Appears in: P1-17.1, P1-17.2

Python interpreter¶

• Meaning: A Python interpreter is the program that reads and executes Python code. It connects Python statements in a file, notebook, terminal, or IDE to actual runtime behavior.
• Why it matters: Python language syntax and the specific interpreter running it are different layers. This distinction helps explain version differences, package availability, virtual environments, and why correct code may fail in the wrong environment.
• Related concepts: interpreter, script, interactive mode
• Core Section: P2-7.3
• Appears in: P2-8.1, P2-10.1

interpreter¶

• Meaning: An interpreter is a program that reads code and executes it directly. In this book it often refers to Python execution, but the broader idea is the runtime layer that turns code into computation and errors.
• Why it matters: Readers need to distinguish writing code from having a program execute that code. The interpreter concept connects terminal commands, scripts, interactive execution, and runtime environment behavior.
• Related concepts: script, interactive mode, runtime
• Core Section: P2-7.3
• Appears in: P2-8.1, P2-10.1

interrupt authority¶

• Meaning: Interrupt authority is the power for a person to stop, cancel, or hold an automated action that is running or about to run. It is a real execution-control power, not just the ability to comment afterward.
• Why it matters: Oversight without the ability to stop execution may fail to prevent harm. This concept helps readers separate formal review from meaningful intervention in automated systems.
• Related concepts: human oversight, approval, stop condition, scope, safety
• Core Section: P1-15.1
• Appears in: P1-15.3

isolation¶

• Meaning: Isolation is a protective design that keeps a tool, process, server, or file access boundary from spreading directly into other systems or sensitive resources.
• Why it matters: Permission and approval alone do not always limit the blast radius of mistakes or attacks. Isolation adds structural boundaries so connected tools and agent capabilities can be used more safely.
• Related concepts: permission, scope, least privilege, security, agent
• Core Section: P1-14.4
• Appears in: P1-15.3

iterable¶

• Meaning: An iterable is an object that can be looped over to produce values one at a time. Lists, strings, dictionaries, files, and generators can all be iterable.
• Why it matters: Iterability is a shared behavior rather than a single data type. It explains why different objects can be used in the same for loop pattern.
• Related concepts: iterator, loop, list
• Core Section: P2-8.4
• Appears in: P2-8.5, P2-10.1

iterator¶

• Meaning: An iterator is the object that actually produces the next value from an iterable. It carries the current progress of iteration and advances one step at a time.
• Why it matters: Iterators explain why some loops are consumed after one pass and why files or generators may need a new iterator to be read again. They make iteration a stateful execution process rather than just syntax.
• Related concepts: iterable, loop, value
• Core Section: P2-8.4
• Appears in: P2-8.5, P2-10.1