public void map(LongWritable key, Text value, Context context)
throws IOException, InterruptedException {
String cur_file =
((FileSplit) context.getInputSplit()).getPath().getParent().getParent().getName();
String train_file = context.getConfiguration().get("train_file");
if (cur_file.equals(train_file)) {
StringTokenizer st = new StringTokenizer(value.toString());
String word = st.nextToken();
String f_id = st.nextToken();
myKey.set(word);
myVal.set(f_id);
context.write(myKey, myVal);
} else {
StringTokenizer st = new StringTokenizer(value.toString());
String word = st.nextToken();
String f_id = st.nextToken();
StringBuilder builder = new StringBuilder(dlt);
while (st.hasMoreTokens()) {
String filename = st.nextToken();
String tf_idf = st.nextToken();
builder.append(filename);
builder.append(dlt);
builder.append(tf_idf);
builder.append("\t");
}
myKey.set(word);
myVal.set(builder.toString());
context.write(myKey, myVal);
}
}
private static void WriteNewCentroids(Context context) throws IOException {
FileSystem fs = FileSystem.get(context.getConfiguration());
Path nextCFile = new Path(context.getConfiguration().get("NEXTCFILE"));
DataOutputStream d = new DataOutputStream(fs.create(nextCFile, false));
BufferedWriter writer = new BufferedWriter(new OutputStreamWriter(d));
for (String centroid : newCentroids) {
writer.write(centroid + "\n");
}
writer.close();
}
protected void setup(Context context) throws IOException, InterruptedException {
FileSystem fs = FileSystem.get(context.getConfiguration());
Path cFile = new Path(context.getConfiguration().get("CFILE"));
DataInputStream d = new DataInputStream(fs.open(cFile));
BufferedReader reader = new BufferedReader(new InputStreamReader(d));
String line;
while ((line = reader.readLine()) != null) {
StringTokenizer tokenizer = new StringTokenizer(line.toString());
if (tokenizer.hasMoreTokens()) {
List<Double> centroid = new ArrayList<Double>(58);
while (tokenizer.hasMoreTokens()) {
centroid.add(Double.parseDouble(tokenizer.nextToken()));
}
centroids.add(centroid);
}
}
k = centroids.size();
}
public void map(LongWritable key, Text value, Context context)
throws IOException, InterruptedException {
Configuration c = context.getConfiguration();
String s = value.toString();
String input[] = s.split(",");
Text outputkey = new Text();
Text outputvalue = new Text();
double result = 0.0;
/* multiplies matrix and vector entry with matching column value */
result = (Double.parseDouble(input[2])) * (vector.get(Long.parseLong(input[1])));
outputkey.set(input[0]);
outputvalue.set(Double.toString(result));
context.write(outputkey, outputvalue);
}
/* called once at the beginning of the task */
public void setup(Context context) throws IOException, InterruptedException {
BufferedReader br = null;
Path[] files = DistributedCache.getLocalCacheFiles(context.getConfiguration());
if (files != null && files.length > 0)
br = new BufferedReader(new FileReader(files[0].toString()));
String line = null;
/* reads the cached file into a hashmap */
try {
while ((line = br.readLine()) != null) {
String input[] = line.split(",");
vector.put(Long.valueOf(input[0]), Double.valueOf(input[1]));
}
} finally {
br.close();
}
}
protected void setup(Context context) {
Configuration config = context.getConfiguration();
inMovies = config.get("inParameter");
}
