Utils
LogicCircuits.Utils
— ModuleModule with general utilities and missing standard library features that could be useful in any Julia project
LogicCircuits.Utils.AbstractBitVector
— ConstantRetro-fitted super type of all bit vectors
LogicCircuits.Utils.CuBitVector
— TypeCustom CUDA version of BitVector (lacking lots of functionality, just a container for now).
LogicCircuits.Utils.Lit
— TypeLiterals are represented as 32-bit signed integers. Positive literals are positive integers identical to their variable. Negative literals are their negations. Integer 0 should not be used to represent literals.
LogicCircuits.Utils.Var
— TypeVariables are represented as 32-bit unsigned integers
Base.eltype
— MethodFind a type that can capture all column values
Base.isdisjoint
— MethodAre the given sets disjoint (no shared elements)?
LogExpFunctions.logsumexp
— MethodMarginalize out dimensions dims
from log-probability tensor
LogicCircuits.Utils.always
— MethodAn array of 100% probabilities for the given element type
LogicCircuits.Utils.bagging_dataset
— MethodReturns an array of DataFrames where each DataFrame is randomly sampled from the original dataset data
LogicCircuits.Utils.batch
— FunctionCreate mini-batches
LogicCircuits.Utils.batch_size
— MethodBatch size of the dataset
LogicCircuits.Utils.bits_per_pixel
— MethodNormalize the given log-likelihood as bits per pixel in data
LogicCircuits.Utils.chunks
— MethodRetrieve chunks of bit vector
LogicCircuits.Utils.eachcol_unweighted
— MethodIterate over columns, excluding the sample weight column
LogicCircuits.Utils.example
— MethodGet the ith example
LogicCircuits.Utils.feature_values
— MethodGet the ith feature values
LogicCircuits.Utils.fully_factorized_log_likelihood
— MethodComputer the per-example log-likelihood of a fully factorized ML model on Bool data
LogicCircuits.Utils.get_bit
— MethodRetrieve the jth bit from a BitVector
chunk
LogicCircuits.Utils.get_weights
— MethodGet the weights from a weighted dataset.
LogicCircuits.Utils.impute
— MethodReturn a copy of Imputed values of X (potentially statistics from another DataFrame)
For example, to impute using same DataFrame:
impute(X; method=:median)
If you want to use another DataFrame to provide imputation statistics:
impute(test_x, train_x; method=:mean)
Supported methods are :median
, :mean
, :one
, :zero
LogicCircuits.Utils.init_array
— MethodAn array of undetermined values (fast) for the given element type
LogicCircuits.Utils.isbatched
— MethodIs the dataset batched?
LogicCircuits.Utils.isbinarydata
— MethodIs the dataset binary?
LogicCircuits.Utils.iscomplete
— MethodIs the data complete (no missing values)?
LogicCircuits.Utils.iscomplete_col
— MethodIs the data column complete (no missing values)?
LogicCircuits.Utils.isfpdata
— MethodIs the dataset consisting of floating point data?
LogicCircuits.Utils.isgpu
— MethodCheck whether data resides on the GPU
LogicCircuits.Utils.issomething
— MethodIs the argument not nothing
?
LogicCircuits.Utils.isweighted
— MethodIs the dataset weighted?
LogicCircuits.Utils.lit2var
— MethodConvert a literal its variable, removing the sign of the literal
LogicCircuits.Utils.ll_per_example
— MethodNormalize the given log-likelihood by the number of examples in data
LogicCircuits.Utils.make_missing_mcar
— Methodmake_missing_mcar(d::DataFrame; keep_prob::Float64=0.8)
Returns a copy of dataframe with making some features missing as MCAR, with keep_prob
as probability of keeping each feature.
LogicCircuits.Utils.marginal_prob
— MethodCompute the marginal prob of each feature in a binary dataset.
LogicCircuits.Utils.never
— MethodAn array of 0% probabilities for the given element type
LogicCircuits.Utils.noop
— MethodFunction that does nothing
LogicCircuits.Utils.num_chunks
— MethodFor binary complete data, how many UInt64
bit strings are needed to store one feature?
LogicCircuits.Utils.num_examples
— Methodnum_examples(df::DataFrame)
Number of examples in data
LogicCircuits.Utils.num_features
— Methodnum_features(df::DataFrame)
Number of features in the data
LogicCircuits.Utils.num_variables
— MethodNumber of variables in the data structure
LogicCircuits.Utils.order_asc
— MethodOrder the arguments in a tuple in ascending order
LogicCircuits.Utils.pushrand!
— MethodPush element
into random position in vectorv
LogicCircuits.Utils.random_sample
— FunctionRandomly draw samples from the dataset with replacement
LogicCircuits.Utils.same_device
— MethodEnsure that x
resides on the same device as data
LogicCircuits.Utils.shuffle_examples
— Methodshuffle_examples(df::DataFrame)
Shuffle the examples in the data
LogicCircuits.Utils.similar!
— MethodReuse a given array if it has the right type and size, otherwise make a new one
LogicCircuits.Utils.soften
— FunctionTurn binary data into floating point data close to 0 and 1.
LogicCircuits.Utils.split_sample_weights
— MethodSplit a weighted dataset into unweighted dataset and its corresponding weights.
LogicCircuits.Utils.subseteq_fast
— MethodReplacement for BitSet.⊆ that does not allocate a new BitSet
LogicCircuits.Utils.threshold
— MethodThreshold a numeric dataset making it binary
LogicCircuits.Utils.to_cpu
— MethodMove data to the CPU
LogicCircuits.Utils.to_gpu
— MethodMove data to the GPU
LogicCircuits.Utils.uniform
— MethodAn array of uniform probabilities
LogicCircuits.Utils.var2lit
— MethodConvert a variable to the corresponding positive literal
LogicCircuits.Utils.variables
— FunctionGet the BitSet
of variables in the data structure
LogicCircuits.Utils.weigh_samples
— MethodCreate a weighted copy of the data set