diff options
| author | Xiangrui Meng <meng@apache.org> | 2014-09-01 05:06:58 +0000 |
|---|---|---|
| committer | Xiangrui Meng <meng@apache.org> | 2014-09-01 05:06:58 +0000 |
| commit | 46d52fbb9be4b5b90a7a1ee9ce3e943156d190b9 (patch) | |
| tree | fc8301fb7661e98f37878c2816e1044979f38617 /mllib/index.md | |
| parent | 1f32de1de92160c203fa8bedaff67cf4bf0d835f (diff) | |
| download | spark-website-46d52fbb9be4b5b90a7a1ee9ce3e943156d190b9.tar.gz spark-website-46d52fbb9be4b5b90a7a1ee9ce3e943156d190b9.tar.bz2 spark-website-46d52fbb9be4b5b90a7a1ee9ce3e943156d190b9.zip | |
update mllib webpage
Diffstat (limited to 'mllib/index.md')
| -rw-r--r-- | mllib/index.md | 22 |
1 files changed, 12 insertions, 10 deletions
diff --git a/mllib/index.md b/mllib/index.md index 9d945bf36..66e62df6c 100644 --- a/mllib/index.md +++ b/mllib/index.md @@ -29,9 +29,9 @@ subproject: MLlib points = spark.textFile(<span class="string">"hdfs://..."</span>)<br/> .<span class="sparkop">map</span>(<span class="closure">parsePoint</span>)<br/> <br/> - model = KMeans.<span class="sparkop">train</span>(points) + model = KMeans.<span class="sparkop">train</span>(points, k=10) </div> - <div class="caption">Calling MLlib in Scala</div> + <div class="caption">Calling MLlib in Python</div> </div> </div> </div> @@ -82,16 +82,18 @@ subproject: MLlib <div class="col-md-4 col-padded"> <h3>Algorithms</h3> <p> - MLlib 0.9 contains the following algorithms: + MLlib 1.1 contains the following algorithms: </p> <ul class="list-narrow"> - <li>K-means clustering with <a href="http://theory.stanford.edu/~sergei/papers/vldb12-kmpar.pdf">K-means|| initialization</a>.</li> - <li>L<sub>1</sub>- and L<sub>2</sub>-regularized <a href="http://en.wikipedia.org/wiki/Linear_regression">linear regression</a>.</li> - <li>L<sub>1</sub>- and L<sub>2</sub>-regularized <a href="http://en.wikipedia.org/wiki/Logistic_regression">logistic regression</a>.</li> - <li><a href="http://www.hpl.hp.com/personal/Robert_Schreiber/papers/2008%20AAIM%20Netflix/netflix_aaim08(submitted).pdf">Alternating least squares</a> collaborative filtering, with explicit - ratings or <a href="http://www2.research.att.com/~yifanhu/PUB/cf.pdf">implicit feedback</a>.</li> - <li><a href="http://en.wikipedia.org/wiki/Naive_Bayes_classifier">Naive Bayes</a> multinomial classification.</li> - <li>Stochastic gradient descent.</li> + <li>linear SVM and logistic regression</li> + <li>classification and regression tree</li> + <li>k-means clustering</li> + <li>recommendation via alternating least squares</li> + <li>singular value decomposition</li> + <li>linear regression with L<sub>1</sub>- and L<sub>2</sub>-regularization</li> + <li>multinomial naive Bayes</li> + <li>basic statistics</li> + <li>feature transformations</li> </ul> <p>Refer to the <a href="{{site.url}}docs/latest/mllib-guide.html">MLlib guide</a> for usage examples.</p> </div> |