In statistics, a sample is a collection of observations or records, where each record is a measurement on one variable, or a vector of measurements on multiple variables for one subject.
In the machine learning and computer science communities, though, “sample” often refers to a single record, as shown in this Python Decision Tree extract. “samples = 3000” in statistical terms means “3000 records.”
In fact there are quite a number of synonyms for “record.” From the machine learning & computer science community, in addition to “sample,” we have “instance,” “example,” and “pattern.” In statistics, we have, as noted, “record” and “observation.” The term “case” is more widely used, and, from the database world we have the term “row.”