{"id":19168830,"url":"https://github.com/nikoshet/spark-mlp","last_synced_at":"2025-08-22T08:03:41.292Z","repository":{"id":198450536,"uuid":"294908636","full_name":"nikoshet/spark-mlp","owner":"nikoshet","description":"Multilayer Perceptron Implementation Using Spark","archived":false,"fork":false,"pushed_at":"2020-11-09T10:07:35.000Z","size":36,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-01-03T23:22:40.763Z","etag":null,"topics":["hdfs","machine-learning","mapreduce","multilayer-perceptron","pyspark","python","spark"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/nikoshet.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2020-09-12T09:06:53.000Z","updated_at":"2020-11-09T10:07:37.000Z","dependencies_parsed_at":null,"dependency_job_id":"a12a9da1-6c6a-483b-ba9b-c47c7bdcb924","html_url":"https://github.com/nikoshet/spark-mlp","commit_stats":null,"previous_names":["nikoshet/spark-mlp"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nikoshet%2Fspark-mlp","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nikoshet%2Fspark-mlp/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nikoshet%2Fspark-mlp/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nikoshet%2Fspark-mlp/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/nikoshet","download_url":"https://codeload.github.com/nikoshet/spark-mlp/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":240250350,"owners_count":19771775,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["hdfs","machine-learning","mapreduce","multilayer-perceptron","pyspark","python","spark"],"created_at":"2024-11-09T09:43:56.990Z","updated_at":"2025-02-22T23:41:09.223Z","avatar_url":"https://github.com/nikoshet.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Multilayer Perceptron Implementation Using Spark And Python\n\n## Dataset\n\nThe dataset is the [Customer Complaint Database](https://catalog.data.gov/dataset/consumer-complaint-database). It includes complains from customers on some financial products and services from 2011 untill today. The data file is a comma-delimited .csv file. \n\n### File Format\n\n\n```\n0 \u003c- date %Y-%m-%d\n1 \u003c- category\n2 \u003c- comment\n```\n\n## Main Goal\n\nThe goal of this project is to use [Spark](https://spark.apache.org/) and [HDFS](https://hadoop.apache.org/docs/r1.2.1/hdfs_design.html) to implement a [Perceptron Classifier](https://en.wikipedia.org/wiki/Multilayer_perceptron) using the [TFIDF metric](https://en.wikipedia.org/wiki/Tf%E2%80%93idf). Both RDDs and Spark Dataframe API were utilized.\n\n## Requirements\n- nltk\n- pyspark\n\n## Usage\n\nAssuming that Spark and HDFS are properly installed and working on our system:\n\n- Upload data file in HDFS\n```\nhadoop fs -put ./customer_complaints.csv hdfs://master:9000/customer_complaints.csv\n```\n\n- Install necessary requirements\n```\npip install -r requirements.txt\n```\n\n- Submit job in a Spark environment\n```\nspark-submit mlp.py\n```\n\n\n## Algorithm\n\n- Clean the data\n- Keep k most common words in all comments\n- Remove less often categories\n- Compute TFIDF metric for each word in the comments \n- Use SparseVector where key is word_index and value is TFIDF metric\n- Transform string labels (categories) in integers\n- Split dataset in train and test set (stratified split)\n- Train a Multilayer Perceptron model\n- Compute accuracy of model on test set\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnikoshet%2Fspark-mlp","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnikoshet%2Fspark-mlp","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnikoshet%2Fspark-mlp/lists"}