DeepSharp.Pipelines
Pipeline-driven design for .NET: declare the whole course from raw data to a validated model once, as one artefact, and replay it. The chain is the enforcement — anything that learns from the data is only offered after you have said how to split, so fitting on everything is not a mistake you can make and be warned about later. A declaration written in C# and one written as a file reach equally far, and each is the other read back. Each column's kind is proposed from what every cell says, for a person to accept or change; a profile names what should not be there; a learner says what it needs where the features are handed over — every feature declared to land between minus one and one, numbers left on their own scale, or categories as places — and a run for it leaves out the steps it has no use for; and a report, declared with the pipeline before any number exists, measures a trained model in the answer's own units, each measure beside what predicting the training rows' average scores.
Activity
- Latest release
- 3d ago
- Total releases
- 17
- Cadence
- ~daily
- Last 12 months
- 17
Details
- License
- MIT
- First release
- Sep 23, 2026
| Version | Released | |
|---|---|---|
0.17.0
minor
| ||
0.16.0
minor
| ||
0.15.0
minor
| ||
0.14.0
minor
| ||
0.13.0
minor
| ||
0.12.0
minor
| ||
0.11.0
minor
| ||
0.10.0
minor
| ||
0.9.0
minor
| ||
0.8.0
minor
| ||
0.7.1
patch
| ||
0.7.0
minor
| ||
0.6.1
minor
| ||
0.5.0
minor
| ||
0.4.0
minor
| ||
0.3.0
minor
| ||
0.2.0
initial
|