MesoNet: code archive navigation
code map for finding things again. PDF filenames can run into the preceding text—check the repo when the boundary looks strange
Repository reference: blizzard-labs/NMCtests. The March 2025 Code Base.pdf is an archived listing containing prototypes, imported examples and reference implementations; it is not a clean statement that every file is the final MesoNet model.
| Archive location | Useful entry points |
|---|---|
base_model/main.py |
evalInstance, loadMNIST, inputEncoding, diehl_norm_weights, train |
base_model/mnistV1.py |
temporal_switch, bipolar events, neuronUpdates, STDP connections |
imgClassifierSDNN/src/dog_filter.py |
dog, dog_filter, CPU/GPU normalization variants |
mnistSTDP-CPU |
evaluation.py, main.py, random_conn.py |
neuron-models |
logistic_test.py, summation_test.py |
How to use this archive: search by the directory and function names above, then compare with the repository revision intended for the experiment. PDF extraction removes indentation and sometimes joins filenames to the preceding code. No source-code line numbers are reconstructed from PDF line positions.
Review questions: which code branch generated each published result, which files are unchanged external examples, and which prototype settings were replaced? A single aggregate PDF is useful for navigation, but does not substitute for a run manifest with commit, environment and configuration.
NMCtests experiment folders