Author: tommaso
Date: Fri Oct 14 07:18:02 2016
New Revision: 1764824
URL: http://svn.apache.org/viewvc?rev=1764824&view=rev
Log:
added apprunner plugin, fixed bug in sRNN, adjusted test
Added:
labs/yay/trunk/core/src/main/java/org/apache/yay/NNRunner.java (with
props)
Modified:
labs/yay/trunk/core/pom.xml
labs/yay/trunk/core/src/main/java/org/apache/yay/RNN.java
labs/yay/trunk/core/src/main/java/org/apache/yay/StackedRNN.java
labs/yay/trunk/core/src/test/java/org/apache/yay/RNNCrossValidationTest.java
Modified: labs/yay/trunk/core/pom.xml
URL:
http://svn.apache.org/viewvc/labs/yay/trunk/core/pom.xml?rev=1764824&r1=1764823&r2=1764824&view=diff
==============================================================================
--- labs/yay/trunk/core/pom.xml (original)
+++ labs/yay/trunk/core/pom.xml Fri Oct 14 07:18:02 2016
@@ -72,6 +72,27 @@
<argLine>-Xmx8g</argLine>
</configuration>
</plugin>
+ <plugin>
+ <groupId>org.codehaus.mojo</groupId>
+ <artifactId>appassembler-maven-plugin</artifactId>
+ <version>1.1.1</version>
+ <executions>
+ <execution>
+ <phase>package</phase>
+ <goals>
+ <goal>assemble</goal>
+ </goals>
+ <configuration>
+ <programs>
+ <program>
+
<mainClass>org.apache.yay.NNRunner</mainClass>
+ <name>nn</name>
+ </program>
+ </programs>
+ </configuration>
+ </execution>
+ </executions>
+ </plugin>
</plugins>
</build>
</project>
\ No newline at end of file
Added: labs/yay/trunk/core/src/main/java/org/apache/yay/NNRunner.java
URL:
http://svn.apache.org/viewvc/labs/yay/trunk/core/src/main/java/org/apache/yay/NNRunner.java?rev=1764824&view=auto
==============================================================================
--- labs/yay/trunk/core/src/main/java/org/apache/yay/NNRunner.java (added)
+++ labs/yay/trunk/core/src/main/java/org/apache/yay/NNRunner.java Fri Oct 14
07:18:02 2016
@@ -0,0 +1,91 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied. See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+package org.apache.yay;
+
+import java.io.IOException;
+import java.nio.file.Files;
+import java.nio.file.Path;
+import java.nio.file.Paths;
+import java.util.Random;
+
+/**
+ * Runner class for different types of neural network
+ */
+public class NNRunner {
+
+ public static void main(String[] args) {
+ if (args.length > 0) {
+ Random random = new Random();
+ String nnType = args[0];
+ switch (nnType) {
+ case "skipgram": {
+ // skipgram network
+ break;
+ }
+ case "recurrent": {
+ // recurrent neural network
+ // e.g. bin/nn recurrent
core/src/test/resources/word2vec/sentences.txt true 100 25 100 stacked
+ float learningRate = 1e-2f;
+ int seqLength = 25;
+ int hiddenLayerSize = 30;
+ int epochs = 20;
+ boolean useChars = true;
+ String text = "";
+
+ if (args.length > 1 && args[1] != null) {
+ Path path = Paths.get(args[1]);
+ try {
+ byte[] bytes = Files.readAllBytes(path);
+ text = new String(bytes);
+ } catch (IOException e) {
+ throw new RuntimeException("could not read from path " + path);
+ }
+ }
+ if (args.length > 2 && args[2] != null) {
+ useChars = Boolean.valueOf(args[2]);
+ }
+ if (args.length > 3 && args[3] != null) {
+ epochs = Integer.valueOf(args[3]);
+ }
+ if (args.length > 4 && args[4] != null) {
+ hiddenLayerSize = Integer.valueOf(args[4]);
+ }
+ if (args.length > 5 && args[5] != null) {
+ seqLength = Integer.valueOf(args[5]);
+ }
+ RNN rnn = new RNN(learningRate, seqLength, hiddenLayerSize, epochs,
text, useChars);
+ if (args.length > 6 && args[6] != null) {
+ if ("stacked".equals(args[6])) {
+ rnn = new StackedRNN(learningRate, seqLength, hiddenLayerSize,
epochs, text, useChars);
+ }
+ }
+
+ rnn.learn();
+ int seed = random.nextInt(rnn.vocabSize);
+ System.out.println(rnn.sample(seed));
+ break;
+ }
+ case "multi": {
+ // multi layer network
+ break;
+ }
+ }
+ }
+ }
+}
Propchange: labs/yay/trunk/core/src/main/java/org/apache/yay/NNRunner.java
------------------------------------------------------------------------------
svn:eol-style = native
Modified: labs/yay/trunk/core/src/main/java/org/apache/yay/RNN.java
URL:
http://svn.apache.org/viewvc/labs/yay/trunk/core/src/main/java/org/apache/yay/RNN.java?rev=1764824&r1=1764823&r2=1764824&view=diff
==============================================================================
--- labs/yay/trunk/core/src/main/java/org/apache/yay/RNN.java (original)
+++ labs/yay/trunk/core/src/main/java/org/apache/yay/RNN.java Fri Oct 14
07:18:02 2016
@@ -26,6 +26,11 @@ import org.nd4j.linalg.api.ops.impl.tran
import org.nd4j.linalg.factory.Nd4j;
import org.nd4j.linalg.ops.transforms.Transforms;
+import java.io.BufferedWriter;
+import java.io.File;
+import java.io.FileWriter;
+import java.io.IOException;
+import java.util.Date;
import java.util.HashMap;
import java.util.HashSet;
import java.util.LinkedList;
@@ -81,7 +86,7 @@ public class RNN {
Set<String> tokens = new HashSet<>(data);
vocabSize = tokens.size();
- System.out.printf("data has %d tokens, %d unique.", data.size(),
vocabSize);
+ System.out.printf("data has %d tokens, %d unique.\n", data.size(),
vocabSize);
charToIx = new HashMap<>();
ixToChar = new HashMap<>();
int i = 0;
@@ -140,7 +145,7 @@ public class RNN {
INDArray targets = getSequence(p + 1);
// sample from the model every now and then
- if (n % 1000 == 0) {
+ if (n % 1000 == 0 && n > 0) {
String txt = sample(inputs.getInt(0));
System.out.printf("\n---\n %s \n----\n", txt);
}
@@ -330,7 +335,7 @@ public class RNN {
@Override
public String toString() {
- return "RNN{" +
+ return getClass().getName() + "{" +
"learningRate=" + learningRate +
", seqLength=" + seqLength +
", hiddenLayerSize=" + hiddenLayerSize +
@@ -342,7 +347,7 @@ public class RNN {
public String getHyperparamsString() {
- return "RNN{" +
+ return getClass().getName() + "{" +
"wxh=" + wxh +
", whh=" + whh +
", why=" + why +
@@ -350,4 +355,20 @@ public class RNN {
", by=" + by +
'}';
}
+
+ public void serialize(String prefix) throws IOException {
+ BufferedWriter bufferedWriter = new BufferedWriter(new FileWriter(new
File(prefix + new Date().toString() + ".csv")));
+ bufferedWriter.write("wxh");
+ bufferedWriter.write(wxh.toString());
+ bufferedWriter.write("whh");
+ bufferedWriter.write(whh.toString());
+ bufferedWriter.write("why");
+ bufferedWriter.write(why.toString());
+ bufferedWriter.write("bh");
+ bufferedWriter.write(bh.toString());
+ bufferedWriter.write("by");
+ bufferedWriter.write(by.toString());
+ bufferedWriter.flush();
+ bufferedWriter.close();
+ }
}
Modified: labs/yay/trunk/core/src/main/java/org/apache/yay/StackedRNN.java
URL:
http://svn.apache.org/viewvc/labs/yay/trunk/core/src/main/java/org/apache/yay/StackedRNN.java?rev=1764824&r1=1764823&r2=1764824&view=diff
==============================================================================
--- labs/yay/trunk/core/src/main/java/org/apache/yay/StackedRNN.java (original)
+++ labs/yay/trunk/core/src/main/java/org/apache/yay/StackedRNN.java Fri Oct 14
07:18:02 2016
@@ -25,6 +25,11 @@ import org.nd4j.linalg.api.ops.impl.tran
import org.nd4j.linalg.factory.Nd4j;
import org.nd4j.linalg.ops.transforms.Transforms;
+import java.io.BufferedWriter;
+import java.io.File;
+import java.io.FileWriter;
+import java.io.IOException;
+import java.util.Date;
import java.util.LinkedList;
import java.util.List;
@@ -104,7 +109,7 @@ public class StackedRNN extends RNN {
INDArray targets = getSequence(p + 1);
// sample from the model every now and then
- if (n % 1000 == 0) {
+ if (n % 1000 == 0 && n > 0) {
String txt = sample(inputs.getInt(0));
System.out.printf("\n---\n %s \n----\n", txt);
}
@@ -195,9 +200,9 @@ public class StackedRNN extends RNN {
if (hs2 == null) {
hs2 = init(inputs.length(), hst2);
}
- hs.putRow(t, hst);
+ hs2.putRow(t, hst2);
- INDArray yst = (wh2y.mmul(hst)).add(by); // unnormalized log
probabilities for next chars
+ INDArray yst = (wh2y.mmul(hst2)).add(by); // unnormalized log
probabilities for next chars
if (ys == null) {
ys = init(inputs.length(), yst);
}
@@ -226,6 +231,7 @@ public class StackedRNN extends RNN {
dby.addi(dy);
INDArray dh2 = wh2y.transpose().mmul(dy).add(dh2Next); // backprop into
h2
INDArray dhraw2 =
(Nd4j.ones(hst2.shape()).sub(hst2).mul(hst2)).mul(dh2); // backprop through
tanh nonlinearity
+// INDArray dhraw2 = Nd4j.getExecutioner().execAndReturn(new
SetRange(hst2, 0, Double.MAX_VALUE));; // backprop through relu nonlinearity
dbh2.addi(dhraw2);
INDArray hst = hs.getRow(t);
@@ -233,6 +239,7 @@ public class StackedRNN extends RNN {
dbh.addi(dh2);
INDArray dh = whh2.transpose().mmul(dh2).add(dhNext); // backprop into h
INDArray dhraw = (Nd4j.ones(hst.shape()).sub(hst).mul(hst)).mul(dh); //
backprop through tanh nonlinearity
+// INDArray dhraw = Nd4j.getExecutioner().execAndReturn(new SetRange(hst,
0, Double.MAX_VALUE));; // backprop through relu nonlinearity
dbh.addi(dhraw);
dWxh.addi(dhraw.mmul(xs.getRow(t)));
@@ -255,6 +262,7 @@ public class StackedRNN extends RNN {
/**
* sample a sequence of integers from the model, using current (hPrev)
memory state, seedIx is seed letter for first time step
*/
+ @Override
public String sample(int seedIx) {
INDArray x = Nd4j.zeros(vocabSize, 1);
@@ -285,4 +293,25 @@ public class StackedRNN extends RNN {
return getSampleString(ixes);
}
+ @Override
+ public void serialize(String prefix) throws IOException {
+ BufferedWriter bufferedWriter = new BufferedWriter(new FileWriter(new
File(prefix + new Date().toString() + ".csv")));
+ bufferedWriter.write("wxh");
+ bufferedWriter.write(wxh.toString());
+ bufferedWriter.write("whh");
+ bufferedWriter.write(whh.toString());
+ bufferedWriter.write("whh2");
+ bufferedWriter.write(whh2.toString());
+ bufferedWriter.write("wh2y");
+ bufferedWriter.write(wh2y.toString());
+ bufferedWriter.write("bh");
+ bufferedWriter.write(bh.toString());
+ bufferedWriter.write("bh2");
+ bufferedWriter.write(bh2.toString());
+ bufferedWriter.write("by");
+ bufferedWriter.write(by.toString());
+ bufferedWriter.flush();
+ bufferedWriter.close();
+ }
+
}
Modified:
labs/yay/trunk/core/src/test/java/org/apache/yay/RNNCrossValidationTest.java
URL:
http://svn.apache.org/viewvc/labs/yay/trunk/core/src/test/java/org/apache/yay/RNNCrossValidationTest.java?rev=1764824&r1=1764823&r2=1764824&view=diff
==============================================================================
---
labs/yay/trunk/core/src/test/java/org/apache/yay/RNNCrossValidationTest.java
(original)
+++
labs/yay/trunk/core/src/test/java/org/apache/yay/RNNCrossValidationTest.java
Fri Oct 14 07:18:02 2016
@@ -19,6 +19,7 @@
package org.apache.yay;
import org.apache.commons.io.IOUtils;
+import org.junit.Before;
import org.junit.Test;
import org.junit.runner.RunWith;
import org.junit.runners.Parameterized;
@@ -39,6 +40,9 @@ public class RNNCrossValidationTest {
private int seqLength;
private int hiddenLayerSize;
private Random r = new Random();
+ private String text;
+ private final int epochs = 2;
+ private List<String> words;
public RNNCrossValidationTest(float learningRate, int seqLength, int
hiddenLayerSize) {
this.learningRate = learningRate;
@@ -46,71 +50,73 @@ public class RNNCrossValidationTest {
this.hiddenLayerSize = hiddenLayerSize;
}
+ @Before
+ public void setUp() throws Exception {
+ InputStream stream =
getClass().getResourceAsStream("/word2vec/abstracts.txt");
+ text = IOUtils.toString(stream);
+ words = Arrays.asList(text.split(" "));
+ stream.close();
+ }
+
@Parameterized.Parameters
public static Collection<Object[]> data() {
return Arrays.asList(new Object[][]{
- {1e-1f, 25, 100},
- {3e-1f, 25, 100},
- {3e-1f, 100, 25},
- {1e-1f, 100, 25},
+ {3e-1f, 50, 5},
+ {3e-1f, 50, 10},
+ {3e-1f, 50, 15},
+ {3e-1f, 50, 25},
+ {3e-1f, 50, 50},
+ {3e-1f, 50, 100},
});
}
@Test
public void testVanillaCharRNNLearn() throws Exception {
- InputStream resourceAsStream =
getClass().getResourceAsStream("/word2vec/sentences.txt");
- String text = IOUtils.toString(resourceAsStream);
- int epochs = 10;
- RNN RNN = new RNN(learningRate, seqLength, hiddenLayerSize, epochs, text);
- checkCorrectWordsRatio(text, RNN);
+ RNN rnn = new RNN(learningRate, seqLength, hiddenLayerSize, epochs, text);
+ evaluate(rnn, true);
+ rnn.serialize("target/crnn-weights-");
}
@Test
public void testVanillaWordRNNLearn() throws Exception {
- InputStream resourceAsStream =
getClass().getResourceAsStream("/word2vec/sentences.txt");
- String text = IOUtils.toString(resourceAsStream);
- int epochs = 100;
- RNN RNN = new RNN(learningRate, seqLength, hiddenLayerSize, epochs, text,
false);
- checkCorrectWordsRatio(text, RNN);
+ RNN rnn = new RNN(learningRate, seqLength, hiddenLayerSize, epochs, text,
false);
+ evaluate(rnn, false);
+ rnn.serialize("target/wrnn-weights-");
}
@Test
public void testStackedCharRNNLearn() throws Exception {
- InputStream resourceAsStream =
getClass().getResourceAsStream("/word2vec/sentences.txt");
- String text = IOUtils.toString(resourceAsStream);
- int epochs = 100;
- RNN RNN = new StackedRNN(learningRate, seqLength, hiddenLayerSize, epochs,
text);
- checkCorrectWordsRatio(text, RNN);
+ RNN rnn = new StackedRNN(learningRate, seqLength, hiddenLayerSize, epochs,
text);
+ evaluate(rnn, true);
+ rnn.serialize("target/scrnn-weights-");
}
@Test
public void testStackedWordRNNLearn() throws Exception {
- InputStream resourceAsStream =
getClass().getResourceAsStream("/word2vec/sentences.txt");
- String text = IOUtils.toString(resourceAsStream);
- int epochs = 100;
- RNN RNN = new StackedRNN(learningRate, seqLength, hiddenLayerSize, epochs,
text, false);
- checkCorrectWordsRatio(text, RNN);
+ RNN rnn = new StackedRNN(learningRate, seqLength, hiddenLayerSize, epochs,
text, false);
+ evaluate(rnn, false);
+ rnn.serialize("target/swrnn-weights-");
}
- private void checkCorrectWordsRatio(String text, RNN RNN) {
- System.out.println(RNN);
- List<String> words = Arrays.asList(text.split(" "));
- RNN.learn();
+ private void evaluate(RNN rnn, boolean checkRatio) {
+ System.out.println(rnn);
+ rnn.learn();
+ double c = 0;
for (int i = 0; i < 10; i++) {
- double c = 0;
- String sample = RNN.sample(r.nextInt(RNN.getVocabSize()));
- String[] sampleWords = sample.split(" ");
- for (String sw : sampleWords) {
- if (words.contains(sw)) {
- c++;
+ String sample = rnn.sample(r.nextInt(rnn.getVocabSize()));
+ if (checkRatio) {
+ String[] sampleWords = sample.split(" ");
+ for (String sw : sampleWords) {
+ if (words.contains(sw)) {
+ c++;
+ }
+ }
+ if (c > 0) {
+ c /= sampleWords.length;
}
}
- if (c > 0) {
- c /= sample.length();
- }
- System.out.println("correct word ratio: " + c);
-
}
+ System.out.println("average correct word ratio: " + (c / 10d));
}
}
\ No newline at end of file
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@labs.apache.org
For additional commands, e-mail: commits-h...@labs.apache.org
