{"api_version":2,"name":"piflow_word_to_vec","version":"1.0","role":"transform","type":"sdk","description":"Train Spark Word2Vec on the input corpus and append the mean token embedding for each document.","icon":{"path":"M3 18h18M5 15l5-6 4 3 5-7"},"inputs":{"source":{"representations":["records"],"formats":[]}},"outputs":{"result":{"view":"records"}},"parameters":{"type":"object","properties":{"column":{"type":"string","description":"Name of the whitespace-tokenized text column.","minLength":1},"vectorSize":{"type":"integer","description":"Number of embedding coordinates from 1 to 256.","minimum":1,"maximum":256}},"required":["column","vectorSize"],"additionalProperties":false},"documentation":{"inputs":{"source":"Arrow records with a string text column."},"outputs":{"result":"Input records and an embedding array for each document."},"parameters":{"column":{"description":"Name of the whitespace-tokenized text column.","help":"Name of the whitespace-tokenized text column.","example":"text"},"vectorSize":{"description":"Number of embedding coordinates from 1 to 256.","help":"Number of embedding coordinates from 1 to 256.","example":4}},"example_job":"name: PiFlow Word2Vec\nsteps:\n- id: read\n  operator: item_source@1.0\n  inputs: {}\n  params:\n    uri: context:/piflow/example.jsonl\n    view: records\n- id: process\n  operator: piflow_word_to_vec@1.0\n  inputs:\n    source: read\n  params:\n    column: text\n    vectorSize: 4\n- id: receive\n  operator: client_sink@1.0\n  inputs:\n    result: process\n"},"runtime":{"image":"sha256:edd926f5a144036f8fbfd7db6771067b44ef00023872fe68552d3419ab549e90","archive":{"sha256":"5b4a94b1342e29dcbdc310b0efa54739b4e221d84005420e9d54e305fdbd4546","size":425330122,"filename":"piflow-ecoflow-spark-runtime.tar.gz"}},"execution":{"executor":"jvm","entrypoint":"org.apache.spark.sql.piflow.WordToVec","config":{}},"source_public":true,"provenance":{"license":"Apache-2.0","source_repository":"https://github.com/cas-bigdatalab/piflow","source_revision":"777361c4e72b2c5457479683916b7c8e635289ff","build_instructions":"Compile src/ using piflow-ecoflow-spark@1.0. See job.yaml for a workflow and fixtures/example.jsonl for example records.","source_archive":{"sha256":"83692af5c6f6e08bee4572fe928b97b17cac98777958f3c9a18dfa704971efbb","size":11257,"filename":"piflow_word_to_vec-sources.zip"},"license_file":{"sha256":"c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4","size":11357,"filename":"LICENSE.txt"}},"artifact":{"sha256":"1857e86f1eaa6abe3cab37238276ad72d6856dd93d92b6948865de94791ac059","size":21647,"filename":"operator.jar"},"categories":["PiFlow","Machine learning"],"trust":{"score":4,"assessed_by":"administrator","reviewed_at_ms":1795080511043},"publication":{"author":"dacp-examples","submitted_at_ms":1795080038325,"approved_at_ms":1795080511043,"last_submitted_at_ms":1795080038325,"last_approved_at_ms":1795080511043},"download_count":4}