Commit graph

  • 285e16c022 Загрузить файлы в «rover-reflex» rover-mod arnee 2026-10-29 00:41:29 +03:00
  • b5d3855c2c Host digit comparison artifact and expose all KNP demos through Caddy main Server 2026-10-21 14:43:13 +00:00
  • 597e40afaf
    Wiki: rewrite as a connected chain of pages with project engineering details wiki-sharedocs Claude 2026-10-19 11:02:28 +00:00
  • 69ad76bde2
    Wiki: remove out-of-context project details, define terms int 2026-10-19 08:42:01 +00:00
  • 917808edda
    Add project wiki built on SharedDocsLib int 2026-10-19 08:26:11 +00:00
  • 4936928a8d Add native AltAI CUDA inference, MNIST and pet demos with experiment reports int 2026-10-16 18:15:00 +03:00
  • 8fe6c3a248 Save state before session shutdown: resume steps for KNP vehicles pipeline int 2026-10-15 11:29:20 +03:00
  • 1b58551d67 Add 2026-10-07 report (KNP build in WSL) int 2026-10-15 11:03:53 +03:00
  • 38cefe8f53 Reorganise reports by day (reports/<date>/), add index; move Claude memory into claude-memory/ int 2026-10-15 10:48:29 +03:00
  • 151e9f0b23 Inference-on-KNP pipeline: altai_infer, offline model building and readout training int 2026-10-15 10:42:35 +03:00
  • ded57ee2f2 All-on-platform object pipeline: KNP program (retina, competitive features, readout), vehicles data, tuning tools int 2026-10-15 08:11:15 +03:00
  • a8ea8beca0 Summary of findings and plan for next week's demo int 2026-10-15 07:13:02 +03:00
  • ee48cb87bd KNP_W_RANGE option; weight range does not explain object gap int 2026-10-15 06:56:14 +03:00
  • 6bfdbf69d3 Delta-rule KNP: augmented MNIST robustness, object runs; report section int 2026-10-15 06:50:34 +03:00
  • 917b94035d Delta-rule KNP results: MNIST 94.8%, cats/dogs 69.1% on validation int 2026-10-15 05:59:50 +03:00
  • 64fc5161e0 Error-driven dopamine (delta-rule) training mode in KNP: network option, isolated backend int 2026-10-15 04:58:42 +03:00
  • 39079670c0 K=1600 neural-rule pipeline result int 2026-10-15 02:32:32 +03:00
  • ba81f0384b Report: seed variation, neural-rule pipeline, page int 2026-10-15 02:12:58 +03:00
  • febc599ff2 End-to-end neural-rule pipeline result; seed-variation script int 2026-10-15 02:09:00 +03:00
  • 1888412f31 Visualizer: stroke-feature network with spiking readout panel int 2026-10-15 01:58:22 +03:00
  • 2f210b16f6 Column limits on learned features; dopamine-trained spiking readout with calibrated decision neuron int 2026-10-15 01:54:37 +03:00
  • ec913f51c1 Online WTA feature learning; configurable active steps in spiking pipeline int 2026-10-15 00:46:14 +03:00
  • 899e0e3601 Night report: objects, spiking readout, punishment int 2026-10-15 00:44:24 +03:00
  • c2ac33767d One-time fresh-test evaluation of object pipelines int 2026-10-15 00:43:38 +03:00
  • 8f126bc29a Exact spiking pipeline simulator, fresh test set, night log int 2026-10-15 00:13:52 +03:00
  • 52d91e974f Cats/dogs v3 learned features for KNP, punishment in cats_dogs, readout variants int 2026-10-14 21:54:09 +03:00
  • cf386f4e36 Log night experiments: spiking readout, learned object features, punishment int 2026-10-14 21:17:39 +03:00
  • 795f75a65b KNP_COL_PUNISH: dopamine punishment of wrong columns; learned k-means features for objects; extra data fetcher int 2026-10-14 21:06:28 +03:00
  • b8ea9dc6ce Spiking readout (off-chip ridge, AltAI-like integer inference) and robustness check int 2026-10-14 20:58:39 +03:00
  • 7add19dcae Log augmentation A and column-size results int 2026-10-14 20:35:03 +03:00
  • ab54a7c8d5 Log coarse-feature results and classifier diagnosis int 2026-10-14 20:10:27 +03:00
  • 334bfefa81 Coarse stroke-feature preset; log fine-feature results int 2026-10-14 19:49:56 +03:00
  • 45880f1d19 Stroke-detector + pooling features prototype, robustness check in feature mode int 2026-10-14 19:26:07 +03:00
  • 27279ab242 Visualizer: shift buttons and all-nets answer line; page with augmented net B int 2026-10-14 19:23:01 +03:00
  • e5d50c8d65 Log augmentation B robustness results int 2026-10-14 19:20:20 +03:00
  • 1722175ac8 Fix resume.sh completion check, log full ON/OFF result, relaunch runs overnight int 2026-10-14 19:17:29 +03:00
  • b34bdae638 Log ON/OFF and robustness results, add resume script int 2026-10-14 19:10:08 +03:00
  • f477a6f588 Augmentation experiment: training-set augmenter and test-time shift voting for MNIST int 2026-10-14 19:07:59 +03:00
  • db4b1d08b8 ON/OFF input coding for mnist-learn: KNP_MNIST_INPUT_SIZE, dataset builder, simulator --pix int 2026-10-14 18:35:27 +03:00
  • 979aa90bb4 Add 2026-10-14 handoff materials, verify accelerated runs int 2026-10-14 18:29:08 +03:00
  • 8f43927575 Initial import: KNP experiments (examples, viz, cats-dogs, reports) int 2026-10-14 18:17:15 +03:00