Skip to content

verified-vectors.md

Latest commit

 

History

History
63 lines (47 loc) · 2.91 KB

verified-vectors.md

File metadata and controls

63 lines (47 loc) · 2.91 KB
Raw

Verified numerical vectors tutorial

This walkthrough is for the Scarp v0.0.5-alpha.1 interpreter. Its source and expected output below are copied without changes from the released vectors fixture. Both language editions are checked against the pinned compiler examples.

1. Create and inspect vectors

Create a UTF-8 file named vectors.srp. This is one complete runnable program, not a collection of independent snippets:

bind values = [1.0, 2.0, 3.0, 4.0];
print(values, len(values), values[0]);
var weights = [0.2, 0.3, 0.5];
bind snapshot = weights;
weights = scale(weights, 2.0);
print(snapshot);
print(weights);
print(sum(values), mean(values));
print(dot(values, [2.0, -1.0, 0.5, 2.0]));
print(mse([1.0, 2.0], [1.0, 4.0]));
print(clamp([-2.0, 0.5, 3.0], 0.0, 1.0));
print(fill(1.5, 3));

Run the program from the folder containing your file:

scarp check vectors.srp
scarp run vectors.srp

The expected stdout of run is exactly:

[1.0, 2.0, 3.0, 4.0] 4 1.0
[0.2, 0.3, 0.5]
[0.4, 0.6, 1.0]
10.0 2.5
9.5
2.0
[0.0, 0.5, 1.0]
[1.5, 1.5, 1.5]

The output matches the compiler's published stdout fixture.

2. Read the operations

  • [1.0, 2.0, 3.0, 4.0] is a numerical vector; len(values) reports its length and values[0] selects its first element.
  • bind snapshot = weights keeps the earlier value, while reassignment through var weights stores the result of scale(weights, 2.0). The earlier snapshot remains unchanged.
  • sum adds elements; mean computes their average. dot combines matching vector positions; mse computes mean squared error for two equal-length vectors.
  • clamp bounds each element between 0.0 and 1.0, while fill(1.5, 3) constructs three identical entries.

3. Experiment without confusing verified results

Change the values in your own copy and repeat scarp check vectors.srp before scarp run vectors.srp. Your modified program will not necessarily have the same output as the fixture above. Scarp checks vector dimensions, and invalid numeric operations can produce diagnostics. Review numeric types and vectors, diagnostics, and runtime budgets.

4. Continue with modules and tiles

The source modules walkthrough combines a tile matrix and a vector through matvec; its complete file contents and output are also checked against the published compiler. Use the verified functions tutorial if function calls are new to you.

::: warning Prerelease compatibility This is a demonstration of the pinned v0.0.5-alpha.1 interpreter, not evidence that these operations are available in older or future releases. Refer to feature status. :::