Aggregate the values of each key with Aggregator.
Aggregate the values of each key with Aggregator. First
each value V
is mapped to A
, then we reduce with a
Semigroup of A
, then finally we present the results as
U
. This could be more powerful and better optimized in some cases.
Aggregate the values of each key, using given combine functions and a neutral "zero value".
Aggregate the values of each key, using given combine functions and a neutral "zero value".
This function can return a different result type, U
, than the type of the values in this
SCollection, V
. Thus, we need one operation for merging a V
into a U
and one operation
for merging two U
's. To avoid memory allocation, both of these functions are allowed to
modify and return their first argument instead of creating a new
U.
For each key, compute the values' data distribution using approximate N
-tiles.
For each key, compute the values' data distribution using approximate N
-tiles.
a new SCollection whose values are Iterable
s of the approximate N
-tiles of
the elements.
Convert this SCollection to a SideInput, mapping key-value pairs of each window to a
Map[key, value]
, to be used with SCollection.withSideInputs.
Convert this SCollection to a SideInput, mapping key-value pairs of each window to a
Map[key, value]
, to be used with SCollection.withSideInputs. It is required that each
key of the input be associated with a single value.
Currently, the resulting map is required to fit into memory.
Convert this SCollection to a SideInput, mapping key-value pairs of each window to a
Map[key, Iterable[value]]
, to be used with SCollection.withSideInputs.
Convert this SCollection to a SideInput, mapping key-value pairs of each window to a
Map[key, Iterable[value]]
, to be used with SCollection.withSideInputs. In contrast to
asMapSideInput, it is not required that the keys in the input collection be unique.
Currently, the resulting map is required to fit into memory.
For each key k in this
or that1
or that2
or that3
, return a resulting SCollection
that contains a tuple with the list of values for that key in this
, that1
, that2
and
that3
.
For each key k in this
or that1
or that2
, return a resulting SCollection that contains
a tuple with the list of values for that key in this
, that1
and that2
.
For each key k in this
or that
, return a resulting SCollection that contains a tuple with
the list of values for that key in this
as well as that
.
Generic function to combine the elements for each key using a custom set of aggregation functions.
Generic function to combine the elements for each key using a custom set of aggregation
functions. Turns an SCollection[(K, V)]
into a result of type SCollection[(K, C)]
, for a
"combined type" C
Note that V
and C
can be different -- for example, one might group an
SCollection of type (Int, Int)
into an SCollection of type (Int, Seq[Int])
. Users provide
three functions:
- createCombiner
, which turns a V
into a C
(e.g., creates a one-element list)
- mergeValue
, to merge a V
into a C
(e.g., adds it to the end of a list)
- mergeCombiners
, to combine two C
's into a single one.
Count approximate number of distinct values for each key in the SCollection.
Count approximate number of distinct values for each key in the SCollection.
the maximum estimation error, which should be in the range
[0.01, 0.5]
.
Count approximate number of distinct values for each key in the SCollection.
Count approximate number of distinct values for each key in the SCollection.
the number of entries in the statistical sample; the higher this number, the
more accurate the estimate will be; should be >= 16
.
Count the number of elements for each key.
Count the number of elements for each key.
a new SCollection of (key, count) pairs
Pass each value in the key-value pair SCollection through a flatMap
function without
changing the keys.
Return an SCollection having its values flattened.
Fold by key with Monoid, which defines the associative
function and "zero value" for V
.
Fold by key with Monoid, which defines the associative
function and "zero value" for V
. This could be more powerful and better optimized in some
cases.
Merge the values for each key using an associative function and a neutral "zero value" which may be added to the result an arbitrary number of times, and must not change the result (e.g., Nil for list concatenation, 0 for addition, or 1 for multiplication.).
Perform a full outer join of this
and that
.
Perform a full outer join of this
and that
. For each element (k, v) in this
, the
resulting SCollection will either contain all pairs (k, (Some(v), Some(w))) for w in that
,
or the pair (k, (Some(v), None)) if no elements in that
have key k. Similarly, for each
element (k, w) in that
, the resulting SCollection will either contain all pairs (k,
(Some(v), Some(w))) for v in this
, or the pair (k, (None, Some(w))) if no elements in
this
have key k.
Group the values for each key in the SCollection into a single sequence.
Group the values for each key in the SCollection into a single sequence. The ordering of elements within each group is not guaranteed, and may even differ each time the resulting SCollection is evaluated.
Note: This operation may be very expensive. If you are grouping in order to perform an aggregation (such as a sum or average) over each key, using PairSCollectionFunctions.aggregateByKey or PairSCollectionFunctions.reduceByKey will provide much better performance.
Note: As currently implemented, groupByKey
must be able to hold all the key-value pairs for
any key in memory. If a key has too many values, it can result in an OutOfMemoryError
.
Alias for cogroup
.
Alias for cogroup
.
Alias for cogroup
.
Perform an inner join by replicating that
to all workers.
Perform an inner join by replicating that
to all workers. The right side should be tiny and
fit in memory.
Perform a left outer join by replicating that
to all workers.
Perform a left outer join by replicating that
to all workers. The right side should be tiny
and fit in memory.
Return an SCollection with the pairs from this
whose keys are in that
.
Return an SCollection containing all pairs of elements with matching keys in this
and
that
.
Return an SCollection containing all pairs of elements with matching keys in this
and
that
. Each pair of elements will be returned as a (k, (v1, v2)) tuple, where (k, v1) is in
this
and (k, v2) is in that
.
Return an SCollection with the keys of each tuple.
Perform a left outer join of this
and that
.
Perform a left outer join of this
and that
. For each element (k, v) in this
, the
resulting SCollection will either contain all pairs (k, (v, Some(w))) for w in that
, or the
pair (k, (v, None)) if no elements in that
have key k.
Pass each value in the key-value pair SCollection through a map
function without changing
the keys.
Return the max of values for each key as defined by the implicit Ordering[T]
.
Return the max of values for each key as defined by the implicit Ordering[T]
.
a new SCollection of (key, maximum value) pairs
Return the min of values for each key as defined by the implicit Ordering[T]
.
Return the min of values for each key as defined by the implicit Ordering[T]
.
a new SCollection of (key, minimum value) pairs
Merge the values for each key using an associative reduce function.
Merge the values for each key using an associative reduce function. This will also perform the merging locally on each mapper before sending results to a reducer, similarly to a "combiner" in MapReduce.
Perform a right outer join of this
and that
.
Perform a right outer join of this
and that
. For each element (k, w) in that
, the
resulting SCollection will either contain all pairs (k, (Some(v), w)) for v in this
, or the
pair (k, (None, w)) if no elements in this
have key k.
Return a subset of this SCollection sampled by key (via stratified sampling).
Return a subset of this SCollection sampled by key (via stratified sampling).
Create a sample of this SCollection using variable sampling rates for different keys as
specified by fractions
, a key to sampling rate map, via simple random sampling with one
pass over the SCollection, to produce a sample of size that's approximately equal to the sum
of math.ceil(numItems * samplingRate)
over all key values.
whether to sample with or without replacement
map of specific keys to sampling rates
SCollection containing the sampled subset
Return a sampled subset of values for each key of this SCollection.
Return a sampled subset of values for each key of this SCollection.
a new SCollection of (key, sampled values) pairs
N to 1 skew-proof flavor of join.
N to 1 skew-proof flavor of join.
Perform a skewed join where some keys on the left hand may be hot, i.e. appear more than
hotKeyThreshold
times. Frequency of a key is estimated with 1 - delta
probability, and the
estimate is within eps * N
of the true frequency.
true frequency <= estimate <= true frequency + eps * N
, where N is the total size of
the left hand side stream so far.
key with hotKeyThreshold
values will be considered hot. Some runners
have inefficient GroupByKey
implementation for groups with more than
10K values. Thus it is recommended to set hotKeyThreshold
to below
10K, keep upper estimation error in mind.
left hand side key com.twitter.algebird.CMSMonoid
// Implicits that enabling CMS-hashing import com.twitter.algebird.CMSHasherImplicits._ val keyAggregator = CMS.aggregator[K](eps, delta, seed) val hotKeyCMS = self.keys.aggregate(keyAggregator) val p = logs.skewedJoin(logMetadata, hotKeyThreshold = 8500, cms=hotKeyCMS)
Read more about CMS: com.twitter.algebird.CMSMonoid.
Make sure to import com.twitter.algebird.CMSHasherImplicits
before using this join.
N to 1 skew-proof flavor of join.
N to 1 skew-proof flavor of join.
Perform a skewed join where some keys on the left hand may be hot, i.e. appear more than
hotKeyThreshold
times. Frequency of a key is estimated with 1 - delta
probability, and the
estimate is within eps * N
of the true frequency.
true frequency <= estimate <= true frequency + eps * N
, where N is the total size of
the left hand side stream so far.
key with hotKeyThreshold
values will be considered hot. Some runners
have inefficient GroupByKey
implementation for groups with more than
10K values. Thus it is recommended to set hotKeyThreshold
to below
10K, keep upper estimation error in mind. If you sample input via
sampleFraction
make sure to adjust hotKeyThreshold
accordingly.
One-sided error bound on the error of each point query, i.e. frequency estimate.
Must lie in (0, 1)
.
A seed to initialize the random number generator used to create the pairwise independent hash functions.
A bound on the probability that a query estimate does not lie within some small
interval (an interval that depends on eps
) around the truth. Must lie in
(0, 1)
.
left side sample fraction. Default is 1.0
- no sampling.
whether to use sampling with replacement, see SCollection.sample.
// Implicits that enabling CMS-hashing import com.twitter.algebird.CMSHasherImplicits._ val p = logs.skewedJoin(logMetadata)
Read more about CMS: com.twitter.algebird.CMSMonoid.
Make sure to import com.twitter.algebird.CMSHasherImplicits
before using this join.
Full outer join for cases when this
is much larger than that
which cannot fit in memory,
but contains a mostly overlapping set of keys as this
, i.e.
Full outer join for cases when this
is much larger than that
which cannot fit in memory,
but contains a mostly overlapping set of keys as this
, i.e. when the intersection of keys
is sparse in this
. A Bloom Filter of keys in that
is used to split this
into 2
partitions. Only those with keys in the filter go through the join and the rest are
concatenated. This is useful for joining historical aggregates with incremental updates.
Read more about Bloom Filter: com.twitter.algebird.BloomFilter.
estimated number of keys in that
false positive probability when computing the overlap
Return an SCollection with the pairs from this
whose keys are not in that
.
Reduce by key with Semigroup.
Reduce by key with Semigroup. This could be more powerful and better optimized than reduceByKey in some cases.
Swap the keys with the values.
Return the top k (largest) values for each key from this SCollection as defined by the
specified implicit Ordering[T]
.
Return the top k (largest) values for each key from this SCollection as defined by the
specified implicit Ordering[T]
.
a new SCollection of (key, top k) pairs
Return an SCollection with the values of each tuple.
Convert this SCollection to an SCollectionWithHotKeyFanout that uses an intermediate node to combine "hot" keys partially before performing the full combine.
Convert this SCollection to an SCollectionWithHotKeyFanout that uses an intermediate node to combine "hot" keys partially before performing the full combine.
constant value for every key
Convert this SCollection to an SCollectionWithHotKeyFanout that uses an intermediate node to combine "hot" keys partially before performing the full combine.
Convert this SCollection to an SCollectionWithHotKeyFanout that uses an intermediate node to combine "hot" keys partially before performing the full combine.
a function from keys to an integer N, where the key will be spread among N intermediate nodes for partial combining. If N is less than or equal to 1, this key will not be sent through an intermediate node.
Extra functions available on SCollections of (key, value) pairs through an implicit conversion.