v0.1.1

• multiclassifier updates
• finalized phonemizer training script. This phonemizer trainer (v0.1.1) will be used for the next goruut release.
  • fix loading of zlib
  • used transition memoization optimalization, use improvement possible
  • remember local minima and avoid them
• parallel: add LoopUntil()

Full Changelog: v0.1.0...v0.1.1

v0.1.0

• Tally4 is now mandatory - battle tested on various datasets
• Infer2 is now mandatory
• Breaking change to majority pooling is recommended, use New2, New will be removed in future major version
• New pretrained models
• Parallel package with thread safe loop

Full Changelog: v0.0.12...v0.1.0

v0.0.12

• add pprofs
• Make Huge dataset less huge
• Updated commands: train_is_alnum, train_is_virus, train_squareroot
• use parallel loops
• Squareroot demo
• feedforward net: Tally4, Infer2, GetBits and GetClasses

Full Changelog: v0.0.11...v0.0.12

v0.0.11

• add ZLIB algo for models
• mnist generalizes on unseen samples

Full Changelog: v0.0.10...v0.0.11

v0.0.10

• avx tuning
• fix multilayer training bug
• phonemizer pprof support

Full Changelog: v0.0.9...v0.0.10

v0.0.9

• improved multi-layer training

Full Changelog: v0.0.8...v0.0.9

v0.0.8

• fix loading from the new format
• branch training
• virus demo updated

Full Changelog: v0.0.7...v0.0.8

v0.0.7

• Phonemizer dataset and demo
• Add string hashes to hash

Full Changelog: v0.0.6...v0.0.7

v0.0.6

• model save/load to lzw compressed json file/reader/writer

Full Changelog: v0.0.5...v0.0.6

v0.0.5

• New model for is_virus dataset
• Fixed maj pooling combiner for 3+ layers

Full Changelog: v0.0.4...v0.0.5