Language Elements
- Statements
  - if
  - if-else
  - for
  - sfor
  - while
- Learn rules
  - =>
  - _=>
  - +=>
  - .=>
  - #=>
  - !=>
- Infix type 1
  - +
  - -
  - .
  - _
  - __
  - :_
- Infix type 2
  - ==
  - !=
  - >=
  - >
  - <=
  - <
  - &&
  - ||
  - ++
  - --
  - **
  - //
  - %%
  - ^^
  - ..
- Miscellaneous elements
  - |>
  - comment
  - label descent
  - |*>
  - |category: *>
  - |_self>
  - (*,*)
  - |__self>
  - |context>
  - supported-ops
  - dump
  - if-then machine
  - if-then operator
- Object types
- Operator types
  - sigmoid
  - numeric
  - simple
  - compound
  - function
  - bracket
  - powered
  - op-sequence
Operators
- Core Operators
  - Normal
    - coeff-sort
    - do-you-know
    - drop
    - how-many
    - ket-sort
    - measure-currency
    - natural-sort
    - normalize
    - pick-elt
    - rescale
    - reverse
    - rotate-left
    - rotate-right
    - sdrop
    - show-many
    - shuffle
    - sidx
    - sreverse
    - srotate-left
    - srotate-right
    - sshuffle
    - weighted-pick-elt
  - Compound
    - drop-above
    - drop-below
    - normalize
    - pick
    - rescale
    - rotate-left
    - rotate-right
    - select
    - srotate-left
    - srotate-right
    - sselect
  - Compound context
    - sort-by
- Sigmoid Operators
  - Normal
    - ReLU
    - clean
    - invert
    - sigmoid-dec
    - sigmoid-inc
  - Compound
- Ket Operators
  - Normal
    - TM-compress-stars
    - TM-ket-hash
    - ceiling
    - floor
    - gmessage
    - is-prime
    - ket-length
    - log
    - pop-float
    - prime-factors
    - print
    - push-float
    - split
    - ssplit
    - to-comma-number
    - to-lower
    - to-upper
  - Compound
    - Gaussian
    - divide-by
    - has-prefix
    - has-suffix
    - hash
    - int-divide-by
    - is-equal
    - is-greater-equal-than
    - is-greater-than
    - is-in-range
    - is-less-equal-than
    - is-less-than
    - minus
    - mod
    - plus
    - random-int
    - remove-prefix
    - remove-suffix
    - round
    - split
    - ssplit
    - times-by
    - to-lower
    - to-upper
  - Context
    - starts-with
- Superposition Operators
  - Normal
  - Compound
    - bar-chart
    - digit2sp
  - Compound context
    - display-grid
    - gtable
    - learn-grid
    - save-as-dot
    - table
    - tidy-table
    - transpose-table
    - unlearn
- Sequence Operators
  - Normal
  - Context
    - compile
  - Compound
    - display-patch
    - random
    - sdisplay-patch
    - seq-arc-encoder
    - sleep
    - smerge
    - spike-merge
    - sprint
    - tensor-product
  - Compound context
- Function Operators
  - Normal
    - 1 parameter
      - TM-extract-text
      - and
      - is-mbr
      - is-subset
      - mbr
      - not-read
      - not-sread
      - or
      - parallel-sum-sequences
      - read
      - sprint
      - sread
      - substring-position
      - xor
    - 2 parameter
      - and
      - intersection
      - is-equal
      - natural-simm
      - or
      - range
      - simm
      - sprint
      - srange
      - strict-simm
      - string-replace
      - sum
      - swrite
      - transform
      - union
      - unscaled-simm
      - write
      - xor
    - 3 parameter
      - arithmetic
      - range
      - srange
      - value-if
    - 4 parameter
  - Context
    - 1 parameter
      - dump
      - gdump
      - op-zip
      - recursive-dump
      - similar-grid
      - unlearn
    - 2 parameter
    - 3 parameter
      - add-learn
      - learn
      - op-for
      - op-if
      - op-sfor
      - seq-learn
      - smap
    - 4 parameter
      - op-for
      - op-sfor
- All Operators
  - Simple
    - ReLU
    - TM-compress-stars
    - TM-ket-hash
    - TM-sequence-hash
    - ceiling
    - clean
    - coeff-sort
    - compile
    - do-you-know
    - drop
    - extract-category
    - extract-head
    - extract-headless
    - extract-tail
    - extract-value
    - floor
    - gmessage
    - how-many
    - invert
    - is-prime
    - ket-length
    - ket-sort
    - log
    - measure-currency
    - natural-sort
    - normalize
    - pick-elt
    - pop-float
    - prime-factors
    - print
    - push-float
    - rank
    - raw-bar-chart
    - rescale
    - reverse
    - rotate-left
    - rotate-right
    - sdrop
    - seq2sp
    - show-many
    - shuffle
    - sidx
    - sigmoid-dec
    - sigmoid-inc
    - smerge
    - sp2seq
    - split
    - sprint
    - sreverse
    - srotate-left
    - srotate-right
    - sshuffle
    - ssplit
    - starts-with
    - to-comma-number
    - to-lower
    - to-upper
    - weighted-pick-elt
  - Compound
    - Gaussian
    - TM-learn-sentences
    - TM-ngram-partition
    - bar-chart
    - borrow-from-context
    - common
    - copy
    - digit2sp
    - display-grid
    - display-patch
    - divide-by
    - drop-above
    - drop-below
    - equal-input
    - find-inverse
    - gtable
    - has-prefix
    - has-suffix
    - hash
    - inherit
    - inherit-path
    - int-divide-by
    - is-equal
    - is-greater-equal-than
    - is-greater-than
    - is-in-range
    - is-less-equal-than
    - is-less-than
    - learn-grid
    - minus
    - mod
    - natural-similar-input
    - normalize
    - not-threshold-filter
    - op-sum
    - op-union
    - patch-map
    - pick
    - plus
    - random
    - random-int
    - rel-kets
    - remove-prefix
    - remove-suffix
    - rescale
    - rotate-left
    - rotate-right
    - round
    - save-as-dot
    - scompress
    - sdisplay-patch
    - select
    - seq-arc-encoder
    - sigmoid-in-range
    - sigmoid-max
    - sigmoid-min
    - similar-input
    - sleep
    - smap
    - smerge
    - sort-by
    - spatch-map
    - spike-merge
    - split
    - sprint
    - srotate-left
    - srotate-right
    - sselect
    - ssplit
    - star-equal-input
    - stransitive
    - strict-similar-input
    - such-that
    - table
    - tensor-product
    - threshold-filter
    - tidy-table
    - times-by
    - to-lower
    - to-upper
    - transitive
    - transpose-table
    - unlearn
    - unscaled-similar-input
  - Function
    - TM-extract-text
    - add-learn
    - and
    - apply
    - arithmetic
    - dump
    - filter
    - find-operators-between
    - find-path-between
    - gdump
    - grid-simm
    - intersection
    - is-equal
    - is-mbr
    - is-subset
    - learn
    - mbr
    - natural-simm
    - not-filter
    - not-read
    - not-sread
    - op-for
    - op-if
    - op-sfor
    - op-zip
    - or
    - parallel-sum-sequences
    - range
    - read
    - recursive-dump
    - seq-learn
    - similar-grid
    - simm
    - smap
    - sprint
    - srange
    - sread
    - strict-simm
    - string-replace
    - substring-position
    - sum
    - swrite
    - transform
    - union
    - unlearn
    - unscaled-simm
    - value-if
    - write
    - xor
Examples
Website
Github

Home

Operator Usage

Name

unlearn

Description

unlearn[op1, op2, ..., opn] sp
unlearn[*] sp
unlearn the given operator(s) for all the kets in sp
if * is used, then unlearn all supported operators, ie, everything
In some cases where an object has a lot of learn rules, it could be slow
The above is the constant operator version, here is the dynamic version:
unlearn(|op: foo-1> + |op: foo-2> + ...) list-of |kets>

Examples

-- load some toy knowledge:
load fred-sam-friends.sw3

-- see what we know:
dump verbose
    ------------------------------------------
    |context> => |Fred Sam friends>

    supported-ops |Fred> => |op: friends> + |op: age>
    friends |Fred> => |Jack> + |Harry> + |Ed> + |Mary> + |Rob> + |Patrick> + |Emma> + |Charlie>
    age |Fred> => |32>

    supported-ops |Sam> => |op: friends> + |op: age>
    friends |Sam> => |Charlie> + |George> + |Emma> + |Jack> + |Robert> + |Frank> + |Julie>
    age |Sam> => |34>
    ------------------------------------------

-- now delete the age operator for Fred:
unlearn[age] |Fred>

-- see what we now know:
dump verbose
    ------------------------------------------
    |context> => |Fred Sam friends>

    supported-ops |Fred> => |op: friends>
    friends |Fred> => |Jack> + |Harry> + |Ed> + |Mary> + |Rob> + |Patrick> + |Emma> + |Charlie>

    supported-ops |Sam> => |op: friends> + |op: age>
    friends |Sam> => |Charlie> + |George> + |Emma> + |Jack> + |Robert> + |Frank> + |Julie>
    age |Sam> => |34>
    ------------------------------------------

-- now delete the friends operator for Fred:
unlearn[friends] |fred>

-- see what we now know:
dump verbose
    ------------------------------------------
    |context> => |Fred Sam friends>

    supported-ops |Fred> => |>

    supported-ops |Sam> => |op: friends> + |op: age>
    friends |Sam> => |Charlie> + |George> + |Emma> + |Jack> + |Robert> + |Frank> + |Julie>
    age |Sam> => |34>
    ------------------------------------------

See also

Operator type

superposition compound context, context function 1