| AIMACore |  |
 |
|
|---|---|---|---|
| AIMASamples |  |
 |
This package is developed by the author as a sample program development of Artificial Intelligence - A Modern Approach 3rd Edition. This is not the official sample program source for the book. The solution here is just the author's own interpretation and does not have any dependency or ratification from any work carried out as part of the official sample code available at: aimacode. Anyone interested in the official version may visit the link provided.
The environment is split into two parts. AIMACore and AIMASamples. AIMACore contains the core libraries and AIMASamples provides examples to test the core library. Although, the detailed documentation has not been developed, the intent is to the keep the essence of the algorithm as sacrosanct as possible. Algorithms are developed as no side effects such that data types can be governed by underlying applications. Keeping this generalization in mind the typical data structures are not mapped to a specific implementation but parametrized. This way a performance sensitive application may use any data structure that is optimized for the specific application. For someone developing her own application can use the AIMACore library only. There is no reason to add the additional test libraries.
This sample program library is built as Julia packages which can be loaded into the Julia runtime easily. For testing purposes, data files are downloaded from the aimacode/aima-data during the build process.
- Anyone interested in contributing to the library is free to submit PRs.
- However, it's advisable to create an "Issue" and describe what you are working on and how that can benefit the library. Duplication of efforts and surprises are an unnecessary overheads to planning. It's suggested to review issues and collaborate with someone already working on an issue than create a parallel contribution for the same.
- Avoid submitting resolution of multiple issues in one PR. Submission of small PRs makes it easier for review.
- Make sure to add elaborate test cases in
AIMASamplesfor every code change submitted. Untested code is hard to validate for correctness.