Re: [Spark Core] Vectorizing very high-dimensional data sourced in long format
Perhaps it can avoid errors(exhausting executor and driver memory) to add random numbers to the entity_id column when you solve the issue by Patrick's way. Daniel Chalef 于2020年10月31日周六 上午12:42写道: > Yes, the resulting matrix would be sparse. Thanks for the suggestion. Will > explore ways of doing this using an agg and UDF. > > On Fri, Oct 30, 2020 at 6:26 AM Patrick McCarthy > wrote: > >> That's a very large vector. Is it sparse? Perhaps you'd have better luck >> performing an aggregate instead of a pivot, and assembling the vector using >> a UDF. >> >> On Thu, Oct 29, 2020 at 10:19 PM Daniel Chalef >> wrote: >> >>> Hello, >>> >>> I have a very large long-format dataframe (several billion rows) that >>> I'd like to pivot and vectorize (using the VectorAssembler), with the aim >>> to reduce dimensionality using something akin to TF-IDF. Once pivoted, the >>> dataframe will have ~130 million columns. >>> >>> The source, long-format schema looks as follows: >>> >>> root >>> |-- entity_id: long (nullable = false) >>> |-- attribute_id: long (nullable = false) >>> |-- event_count: integer (nullable = true) >>> >>> Pivoting as per the following fails, exhausting executor and driver >>> memory. I am unsure whether increasing memory limits would be successful >>> here as my sense is that pivoting and then using a VectorAssembler isn't >>> the right approach to solving this problem. >>> >>> wide_frame = ( >>> long_frame.groupBy("entity_id") >>> .pivot("attribute_id") >>> .agg(F.first("event_count")) >>> ) >>> >>> Are there other Spark patterns that I should attempt in order to achieve >>> my end goal of a vector of attributes for every entity? >>> >>> Thanks, Daniel >>> >> >> >> -- >> >> >> *Patrick McCarthy * >> >> Senior Data Scientist, Machine Learning Engineering >> >> Dstillery >> >> 470 Park Ave South, 17th Floor, NYC 10016 >> >
Re: [Spark Core] Vectorizing very high-dimensional data sourced in long format
Yes, the resulting matrix would be sparse. Thanks for the suggestion. Will explore ways of doing this using an agg and UDF. On Fri, Oct 30, 2020 at 6:26 AM Patrick McCarthy wrote: > That's a very large vector. Is it sparse? Perhaps you'd have better luck > performing an aggregate instead of a pivot, and assembling the vector using > a UDF. > > On Thu, Oct 29, 2020 at 10:19 PM Daniel Chalef > wrote: > >> Hello, >> >> I have a very large long-format dataframe (several billion rows) that I'd >> like to pivot and vectorize (using the VectorAssembler), with the aim to >> reduce dimensionality using something akin to TF-IDF. Once pivoted, the >> dataframe will have ~130 million columns. >> >> The source, long-format schema looks as follows: >> >> root >> |-- entity_id: long (nullable = false) >> |-- attribute_id: long (nullable = false) >> |-- event_count: integer (nullable = true) >> >> Pivoting as per the following fails, exhausting executor and driver >> memory. I am unsure whether increasing memory limits would be successful >> here as my sense is that pivoting and then using a VectorAssembler isn't >> the right approach to solving this problem. >> >> wide_frame = ( >> long_frame.groupBy("entity_id") >> .pivot("attribute_id") >> .agg(F.first("event_count")) >> ) >> >> Are there other Spark patterns that I should attempt in order to achieve >> my end goal of a vector of attributes for every entity? >> >> Thanks, Daniel >> > > > -- > > > *Patrick McCarthy * > > Senior Data Scientist, Machine Learning Engineering > > Dstillery > > 470 Park Ave South, 17th Floor, NYC 10016 >
Re: [Spark Core] Vectorizing very high-dimensional data sourced in long format
That's a very large vector. Is it sparse? Perhaps you'd have better luck performing an aggregate instead of a pivot, and assembling the vector using a UDF. On Thu, Oct 29, 2020 at 10:19 PM Daniel Chalef wrote: > Hello, > > I have a very large long-format dataframe (several billion rows) that I'd > like to pivot and vectorize (using the VectorAssembler), with the aim to > reduce dimensionality using something akin to TF-IDF. Once pivoted, the > dataframe will have ~130 million columns. > > The source, long-format schema looks as follows: > > root > |-- entity_id: long (nullable = false) > |-- attribute_id: long (nullable = false) > |-- event_count: integer (nullable = true) > > Pivoting as per the following fails, exhausting executor and driver > memory. I am unsure whether increasing memory limits would be successful > here as my sense is that pivoting and then using a VectorAssembler isn't > the right approach to solving this problem. > > wide_frame = ( > long_frame.groupBy("entity_id") > .pivot("attribute_id") > .agg(F.first("event_count")) > ) > > Are there other Spark patterns that I should attempt in order to achieve > my end goal of a vector of attributes for every entity? > > Thanks, Daniel > -- *Patrick McCarthy * Senior Data Scientist, Machine Learning Engineering Dstillery 470 Park Ave South, 17th Floor, NYC 10016