List of usage examples for org.apache.hadoop.io DefaultStringifier toString
@Override public String toString(T obj) throws IOException
From source file:co.nubetech.apache.hadoop.mapred.DBQueryInputFormat.java
License:Apache License
/** * Note that the "orderBy" column is called the "splitBy" in this version. * We reuse the same field, but it's not strictly ordering it -- just * partitioning the results.//from ww w. j av a 2 s . c o m */ public static void setInput(Job job, String tableName, String conditions, String splitBy, ArrayList params, String... fieldNames) throws IOException { DBInputFormat.setInput(job, GenericDBWritable.class, tableName, conditions, splitBy, fieldNames); if (params != null) { DefaultStringifier<ArrayList> stringifier = new DefaultStringifier<ArrayList>(job.getConfiguration(), ArrayList.class); job.getConfiguration().set(HIHOConf.QUERY_PARAMS, stringifier.toString(params)); logger.debug("Converted params and saved them into config"); } job.setInputFormatClass(DBQueryInputFormat.class); }
From source file:co.nubetech.apache.hadoop.mapred.DBQueryInputFormat.java
License:Apache License
/** * setInput() takes a custom query and a separate "bounding query" to use * instead of the custom "count query" used by DBInputFormat. *///from w w w . j a va2 s .c o m public static void setInput(JobConf job, String inputQuery, String inputBoundingQuery, ArrayList params) throws IOException { DBInputFormat.setInput(job, GenericDBWritable.class, inputQuery, ""); if (inputBoundingQuery != null) { job.set(DBConfiguration.INPUT_BOUNDING_QUERY, inputBoundingQuery); } if (params != null) { DefaultStringifier<ArrayList> stringifier = new DefaultStringifier<ArrayList>(job, ArrayList.class); job.set(HIHOConf.QUERY_PARAMS, stringifier.toString(params)); logger.debug("Converted params and saved them into config"); } job.setInputFormat(DBQueryInputFormat.class); }
From source file:co.nubetech.hiho.mapreduce.lib.db.DBQueryInputFormat.java
License:Apache License
/** * setInput() takes a custom query and a separate "bounding query" to use * instead of the custom "count query" used by DBInputFormat. */// w w w . j a va 2 s . co m public static void setInput(Job job, String inputQuery, String inputBoundingQuery, ArrayList params) throws IOException { DBInputFormat.setInput(job, GenericDBWritable.class, inputQuery, ""); if (inputBoundingQuery != null) { job.getConfiguration().set(DBConfiguration.INPUT_BOUNDING_QUERY, inputBoundingQuery); } if (params != null) { DefaultStringifier<ArrayList> stringifier = new DefaultStringifier<ArrayList>(job.getConfiguration(), ArrayList.class); job.getConfiguration().set(HIHOConf.QUERY_PARAMS, stringifier.toString(params)); logger.debug("Converted params and saved them into config"); } job.setInputFormatClass(DBQueryInputFormat.class); }
From source file:org.apache.mahout.avro.text.mapred.WikipediaAvroDocumentMapper.java
License:Apache License
@Override public void configure(JobConf job) { try {/*from ww w .ja v a 2 s.c o m*/ if (inputCategories == null) { Set<String> newCategories = new HashSet<String>(); DefaultStringifier<Set<String>> setStringifier = new DefaultStringifier<Set<String>>(job, GenericsUtil.getClass(newCategories)); String categoriesStr = setStringifier.toString(newCategories); categoriesStr = job.get("wikipedia.categories", categoriesStr); inputCategories = setStringifier.fromString(categoriesStr); } exactMatchOnly = job.getBoolean("exact.match.only", false); all = job.getBoolean("all.files", true); } catch (IOException ex) { throw new IllegalStateException(ex); } log.info("Configure: Input Categories size: " + inputCategories.size() + " All: " + all + " Exact Match: " + exactMatchOnly); }
From source file:org.apache.mahout.avro.text.mapred.WikipediaToAvroDocuments.java
License:Apache License
/** * Run the job//from ww w . j a v a 2 s .co m * * @param input * the input pathname String * @param output * the output pathname String * @param catFile * the file containing the Wikipedia categories * @param exactMatchOnly * if true, then the Wikipedia category must match exactly instead of * simply containing the category string * @param all * if true select all categories */ public static int runJob(String input, String output, String catFile, boolean exactMatchOnly, boolean all) throws IOException { JobClient client = new JobClient(); JobConf conf = new JobConf(WikipediaToAvroDocuments.class); if (log.isInfoEnabled()) { log.info("Input: " + input + " Out: " + output + " Categories: " + catFile + " All Files: " + all); } Path inPath = new Path(input); Path outPath = new Path(output); FileInputFormat.setInputPaths(conf, inPath); FileOutputFormat.setOutputPath(conf, outPath); //AvroOutputFormat.setClass(conf, AvroDocument.class); //AvroOutputFormat.setSchema(conf, AvroDocument._SCHEMA); conf.set("xmlinput.start", "<page>"); conf.set("xmlinput.end", "</page>"); conf.setOutputKeyClass(Text.class); conf.setOutputValueClass(AvroDocument.class); conf.setBoolean("exact.match.only", exactMatchOnly); conf.setBoolean("all.files", all); conf.setMapperClass(WikipediaAvroDocumentMapper.class); conf.setInputFormat(XmlInputFormat.class); conf.setReducerClass(IdentityReducer.class); conf.setOutputFormat(AvroOutputFormat.class); AvroOutputFormat.setAvroOutputClass(conf, AvroDocument.class); FileSystem dfs = FileSystem.get(outPath.toUri(), conf); if (dfs.exists(outPath)) { dfs.delete(outPath, true); } Set<String> categories = new HashSet<String>(); if (catFile.equals("") == false) { for (String line : new FileLineIterable(new File(catFile))) { categories.add(line.trim().toLowerCase()); } } DefaultStringifier<Set<String>> setStringifier = new DefaultStringifier<Set<String>>(conf, GenericsUtil.getClass(categories)); String categoriesStr = setStringifier.toString(categories); conf.set("wikipedia.categories", categoriesStr); client.setConf(conf); RunningJob job = JobClient.runJob(conf); job.waitForCompletion(); return job.isSuccessful() ? 1 : 0; }
From source file:org.apache.mahout.classifier.bayes.BayesThetaNormalizerDriver.java
License:Apache License
/** * Run the job//from ww w . j a v a 2 s . c om * * @param input the input pathname String * @param output the output pathname String */ public static void runJob(String input, String output) throws IOException { JobClient client = new JobClient(); JobConf conf = new JobConf(BayesThetaNormalizerDriver.class); conf.setOutputKeyClass(Text.class); conf.setOutputValueClass(DoubleWritable.class); FileInputFormat.addInputPath(conf, new Path(output + "/trainer-tfIdf/trainer-tfIdf")); Path outPath = new Path(output + "/trainer-thetaNormalizer"); FileOutputFormat.setOutputPath(conf, outPath); conf.setNumMapTasks(100); //conf.setNumReduceTasks(1); conf.setMapperClass(BayesThetaNormalizerMapper.class); conf.setInputFormat(SequenceFileInputFormat.class); conf.setCombinerClass(BayesThetaNormalizerReducer.class); conf.setReducerClass(BayesThetaNormalizerReducer.class); conf.setOutputFormat(SequenceFileOutputFormat.class); conf.set("io.serializations", "org.apache.hadoop.io.serializer.JavaSerialization,org.apache.hadoop.io.serializer.WritableSerialization"); // Dont ever forget this. People should keep track of how hadoop conf parameters and make or break a piece of code FileSystem dfs = FileSystem.get(outPath.toUri(), conf); if (dfs.exists(outPath)) { dfs.delete(outPath, true); } Path Sigma_kFiles = new Path(output + "/trainer-weights/Sigma_k/*"); Map<String, Double> labelWeightSum = SequenceFileModelReader.readLabelSums(dfs, Sigma_kFiles, conf); DefaultStringifier<Map<String, Double>> mapStringifier = new DefaultStringifier<Map<String, Double>>(conf, GenericsUtil.getClass(labelWeightSum)); String labelWeightSumString = mapStringifier.toString(labelWeightSum); log.info("Sigma_k for Each Label"); Map<String, Double> c = mapStringifier.fromString(labelWeightSumString); log.info("{}", c); conf.set("cnaivebayes.sigma_k", labelWeightSumString); Path sigma_kSigma_jFile = new Path(output + "/trainer-weights/Sigma_kSigma_j/*"); double sigma_jSigma_k = SequenceFileModelReader.readSigma_jSigma_k(dfs, sigma_kSigma_jFile, conf); DefaultStringifier<Double> stringifier = new DefaultStringifier<Double>(conf, Double.class); String sigma_jSigma_kString = stringifier.toString(sigma_jSigma_k); log.info("Sigma_kSigma_j for each Label and for each Features"); double retSigma_jSigma_k = stringifier.fromString(sigma_jSigma_kString); log.info("{}", retSigma_jSigma_k); conf.set("cnaivebayes.sigma_jSigma_k", sigma_jSigma_kString); Path vocabCountFile = new Path(output + "/trainer-tfIdf/trainer-vocabCount/*"); double vocabCount = SequenceFileModelReader.readVocabCount(dfs, vocabCountFile, conf); String vocabCountString = stringifier.toString(vocabCount); log.info("Vocabulary Count"); conf.set("cnaivebayes.vocabCount", vocabCountString); double retvocabCount = stringifier.fromString(vocabCountString); log.info("{}", retvocabCount); client.setConf(conf); JobClient.runJob(conf); }
From source file:org.apache.mahout.classifier.bayes.BayesThetaNormalizerMapper.java
License:Apache License
@Override public void configure(JobConf job) { try {/* w w w . j a v a 2 s . c o m*/ if (labelWeightSum == null) { labelWeightSum = new HashMap<String, Double>(); DefaultStringifier<Map<String, Double>> mapStringifier = new DefaultStringifier<Map<String, Double>>( job, GenericsUtil.getClass(labelWeightSum)); String labelWeightSumString = mapStringifier.toString(labelWeightSum); labelWeightSumString = job.get("cnaivebayes.sigma_k", labelWeightSumString); labelWeightSum = mapStringifier.fromString(labelWeightSumString); DefaultStringifier<Double> stringifier = new DefaultStringifier<Double>(job, GenericsUtil.getClass(sigma_jSigma_k)); String sigma_jSigma_kString = stringifier.toString(sigma_jSigma_k); sigma_jSigma_kString = job.get("cnaivebayes.sigma_jSigma_k", sigma_jSigma_kString); sigma_jSigma_k = stringifier.fromString(sigma_jSigma_kString); String vocabCountString = stringifier.toString(vocabCount); vocabCountString = job.get("cnaivebayes.vocabCount", vocabCountString); vocabCount = stringifier.fromString(vocabCountString); } } catch (IOException ex) { log.warn(ex.toString(), ex); } }
From source file:org.apache.mahout.classifier.bayes.common.BayesFeatureDriver.java
License:Apache License
/** * Run the job//from w w w . j a v a 2s .c o m * * @param input the input pathname String * @param output the output pathname String */ public static void runJob(String input, String output, int gramSize) throws IOException { JobClient client = new JobClient(); JobConf conf = new JobConf(BayesFeatureDriver.class); conf.setOutputKeyClass(Text.class); conf.setOutputValueClass(DoubleWritable.class); FileInputFormat.setInputPaths(conf, new Path(input)); Path outPath = new Path(output); FileOutputFormat.setOutputPath(conf, outPath); conf.setNumMapTasks(100); //conf.setNumReduceTasks(1); conf.setMapperClass(BayesFeatureMapper.class); conf.setInputFormat(KeyValueTextInputFormat.class); conf.setCombinerClass(BayesFeatureReducer.class); conf.setReducerClass(BayesFeatureReducer.class); conf.setOutputFormat(BayesFeatureOutputFormat.class); conf.set("io.serializations", "org.apache.hadoop.io.serializer.JavaSerialization,org.apache.hadoop.io.serializer.WritableSerialization"); // Dont ever forget this. People should keep track of how hadoop conf parameters and make or break a piece of code FileSystem dfs = FileSystem.get(outPath.toUri(), conf); if (dfs.exists(outPath)) { dfs.delete(outPath, true); } DefaultStringifier<Integer> intStringifier = new DefaultStringifier<Integer>(conf, Integer.class); String gramSizeString = intStringifier.toString(gramSize); log.info("{}", intStringifier.fromString(gramSizeString)); conf.set("bayes.gramSize", gramSizeString); client.setConf(conf); JobClient.runJob(conf); }
From source file:org.apache.mahout.classifier.bayes.common.BayesFeatureMapper.java
License:Apache License
@Override public void configure(JobConf job) { try {/*from w w w .j ava2 s . c o m*/ DefaultStringifier<Integer> intStringifier = new DefaultStringifier<Integer>(job, Integer.class); String gramSizeString = intStringifier.toString(gramSize); gramSizeString = job.get("bayes.gramSize", gramSizeString); gramSize = intStringifier.fromString(gramSizeString); } catch (IOException ex) { log.warn(ex.toString(), ex); } }
From source file:org.apache.mahout.classifier.bayes.common.BayesTfIdfDriver.java
License:Apache License
/** * Run the job//from w ww . j av a 2 s .co m * * @param input the input pathname String * @param output the output pathname String */ public static void runJob(String input, String output) throws IOException { JobClient client = new JobClient(); JobConf conf = new JobConf(BayesTfIdfDriver.class); conf.setOutputKeyClass(Text.class); conf.setOutputValueClass(DoubleWritable.class); FileInputFormat.addInputPath(conf, new Path(output + "/trainer-termDocCount")); FileInputFormat.addInputPath(conf, new Path(output + "/trainer-wordFreq")); FileInputFormat.addInputPath(conf, new Path(output + "/trainer-featureCount")); Path outPath = new Path(output + "/trainer-tfIdf"); FileOutputFormat.setOutputPath(conf, outPath); conf.setNumMapTasks(100); conf.setMapperClass(BayesTfIdfMapper.class); conf.setInputFormat(SequenceFileInputFormat.class); conf.setCombinerClass(BayesTfIdfReducer.class); conf.setReducerClass(BayesTfIdfReducer.class); conf.setOutputFormat(BayesTfIdfOutputFormat.class); conf.set("io.serializations", "org.apache.hadoop.io.serializer.JavaSerialization,org.apache.hadoop.io.serializer.WritableSerialization"); // Dont ever forget this. People should keep track of how hadoop conf parameters and make or break a piece of code FileSystem dfs = FileSystem.get(outPath.toUri(), conf); if (dfs.exists(outPath)) { dfs.delete(outPath, true); } Path interimFile = new Path(output + "/trainer-docCount/part-*"); Map<String, Double> labelDocumentCounts = SequenceFileModelReader.readLabelDocumentCounts(dfs, interimFile, conf); DefaultStringifier<Map<String, Double>> mapStringifier = new DefaultStringifier<Map<String, Double>>(conf, GenericsUtil.getClass(labelDocumentCounts)); String labelDocumentCountString = mapStringifier.toString(labelDocumentCounts); log.info("Counts of documents in Each Label"); Map<String, Double> c = mapStringifier.fromString(labelDocumentCountString); log.info("{}", c); conf.set("cnaivebayes.labelDocumentCounts", labelDocumentCountString); client.setConf(conf); JobClient.runJob(conf); }