index

README.md

@@ -1,27 +1,46 @@
-![alt text](https://github.com/jhfjhfj1/autokeras/blob/docs/logo.png?raw=true)
-# Welcome to Auto-Keras
-
+<img src="https://github.com/jhfjhfj1/autokeras/blob/docs/logo.png?raw=true" alt="drawing" width="400px"/>
 
 [![Build Status](https://travis-ci.org/jhfjhfj1/autokeras.svg?branch=master)](https://travis-ci.org/jhfjhfj1/autokeras)
 [![Coverage Status](https://coveralls.io/repos/github/jhfjhfj1/autokeras/badge.svg?branch=master)](https://coveralls.io/github/jhfjhfj1/autokeras?branch=master)
 
-This is a automated machine learning (AutoML) package based on Keras. 
+This is a automated machine learning (AutoML) package based on Keras.
 It aims at automatically search for the architecture and hyperparameters for deep learning models.
 The ultimate goal for this project is for domain experts in fields other than computer science or machine learning
 to use deep learning models conveniently.
 
+To install the package please use the commend as follows:
+
+    pip install autokeras
+
 Here is a short example for using the package.
 
-    
+
     import autokeras as ak
-    
+
     (x_train, y_train), (x_test, y_test) = mnist.load_data()
     clf = ak.ImageClassifier()
     clf.fit(x_train, y_train)
     results = clf.predict(x_test)
 
 For the repository on GitHub visit [Auto-Keras on GitHub](https://github.com/jhfjhfj1/autokeras).
 
+If you use Auto-Keras in a scientific publication, we would appreciate references to the following paper:
+
+Efficient Neural Architecture Search with Network Morphism.
+Haifeng Jin, Qingquan Song, Xia Hu.
+[arXiv:1806.10282](https://arxiv.org/abs/1806.10282).
+
+Biblatex entry:
+
+    @online{jin2018efficient,
+      author       = {Haifeng Jin and Qingquan Song and Xia Hu},
+      title        = {Efficient Neural Architecture Search with Network Morphism},
+      date         = {2018-06-27},
+      year         = {2018},
+      eprintclass  = {cs.LG},
+      eprinttype   = {arXiv},
+      eprint       = {cs.LG/1806.10282},
+    }
 
 ### About
 

docs/index.html

@@ -52,9 +52,6 @@
     <a class="current" href=".">Home</a>
     <ul class="subnav">
 
-    <li class="toctree-l2"><a href="#welcome-to-auto-keras">Welcome to Auto-Keras</a></li>
-
-
     </ul>
 	    </li>
 
@@ -126,10 +123,10 @@
           <div role="main">
             <div class="section">
 
-                <h1 id="welcome-to-auto-keras">Welcome to Auto-Keras</h1>
+                <p><img src="https://github.com/jhfjhfj1/autokeras/blob/docs/logo.png?raw=true" alt="drawing" width="400px"/></p>
 <p><a href="https://travis-ci.org/jhfjhfj1/autokeras"><img alt="Build Status" src="https://travis-ci.org/jhfjhfj1/autokeras.svg?branch=master" /></a>
 <a href="https://coveralls.io/github/jhfjhfj1/autokeras?branch=master"><img alt="Coverage Status" src="https://coveralls.io/repos/github/jhfjhfj1/autokeras/badge.svg?branch=master" /></a></p>
-<p>This is a automated machine learning (AutoML) package based on Keras. 
+<p>This is a automated machine learning (AutoML) package based on Keras.
 It aims at automatically search for the architecture and hyperparameters for deep learning models.
 The ultimate goal for this project is for domain experts in fields other than computer science or machine learning
 to use deep learning models conveniently.</p>
@@ -209,5 +206,5 @@ <h1 id="welcome-to-auto-keras">Welcome to Auto-Keras</h1>
 
 <!--
 MkDocs version : 0.17.4
-Build Date UTC : 2018-07-02 02:25:40
+Build Date UTC : 2018-07-04 21:11:41
 -->

docs/search/search_index.json

@@ -2,14 +2,9 @@
     "docs": [
         {
             "location": "/",
-            "text": "Welcome to Auto-Keras\n\n\n\n\n\n\nThis is a automated machine learning (AutoML) package based on Keras. \nIt aims at automatically search for the architecture and hyperparameters for deep learning models.\nThe ultimate goal for this project is for domain experts in fields other than computer science or machine learning\nto use deep learning models conveniently.\n\n\nTo install the package please use the commend as follows:\n\n\npip install autokeras\n\n\n\nHere is a short example for using the package.\n\n\nimport autokeras as ak\n\n(x_train, y_train), (x_test, y_test) = mnist.load_data()\nclf = ak.ImageClassifier()\nclf.fit(x_train, y_train)\nresults = clf.predict(x_test)\n\n\n\nFor the repository on GitHub visit \nAuto-Keras on GitHub\n.\n\n\nIf you use Auto-Keras in a scientific publication, we would appreciate references to the following paper:\n\n\nEfficient Neural Architecture Search with Network Morphism.\nHaifeng Jin, Qingquan Song, Xia Hu.\n\narXiv:1806.10282\n.\n\n\nBiblatex entry:\n\n\n@online{jin2018efficient,\n  author       = {Haifeng Jin and Qingquan Song and Xia Hu},\n  title        = {Efficient Neural Architecture Search with Network Morphism},\n  date         = {2018-06-27},\n  year         = {2018},\n  eprintclass  = {cs.LG},\n  eprinttype   = {arXiv},\n  eprint       = {cs.LG/1806.10282},\n}",
+            "text": "This is a automated machine learning (AutoML) package based on Keras.\nIt aims at automatically search for the architecture and hyperparameters for deep learning models.\nThe ultimate goal for this project is for domain experts in fields other than computer science or machine learning\nto use deep learning models conveniently.\n\n\nTo install the package please use the commend as follows:\n\n\npip install autokeras\n\n\n\nHere is a short example for using the package.\n\n\nimport autokeras as ak\n\n(x_train, y_train), (x_test, y_test) = mnist.load_data()\nclf = ak.ImageClassifier()\nclf.fit(x_train, y_train)\nresults = clf.predict(x_test)\n\n\n\nFor the repository on GitHub visit \nAuto-Keras on GitHub\n.\n\n\nIf you use Auto-Keras in a scientific publication, we would appreciate references to the following paper:\n\n\nEfficient Neural Architecture Search with Network Morphism.\nHaifeng Jin, Qingquan Song, Xia Hu.\n\narXiv:1806.10282\n.\n\n\nBiblatex entry:\n\n\n@online{jin2018efficient,\n  author       = {Haifeng Jin and Qingquan Song and Xia Hu},\n  title        = {Efficient Neural Architecture Search with Network Morphism},\n  date         = {2018-06-27},\n  year         = {2018},\n  eprintclass  = {cs.LG},\n  eprinttype   = {arXiv},\n  eprint       = {cs.LG/1806.10282},\n}",
             "title": "Home"
         },
-        {
-            "location": "/#welcome-to-auto-keras",
-            "text": "This is a automated machine learning (AutoML) package based on Keras. \nIt aims at automatically search for the architecture and hyperparameters for deep learning models.\nThe ultimate goal for this project is for domain experts in fields other than computer science or machine learning\nto use deep learning models conveniently.  To install the package please use the commend as follows:  pip install autokeras  Here is a short example for using the package.  import autokeras as ak\n\n(x_train, y_train), (x_test, y_test) = mnist.load_data()\nclf = ak.ImageClassifier()\nclf.fit(x_train, y_train)\nresults = clf.predict(x_test)  For the repository on GitHub visit  Auto-Keras on GitHub .  If you use Auto-Keras in a scientific publication, we would appreciate references to the following paper:  Efficient Neural Architecture Search with Network Morphism.\nHaifeng Jin, Qingquan Song, Xia Hu. arXiv:1806.10282 .  Biblatex entry:  @online{jin2018efficient,\n  author       = {Haifeng Jin and Qingquan Song and Xia Hu},\n  title        = {Efficient Neural Architecture Search with Network Morphism},\n  date         = {2018-06-27},\n  year         = {2018},\n  eprintclass  = {cs.LG},\n  eprinttype   = {arXiv},\n  eprint       = {cs.LG/1806.10282},\n}",
-            "title": "Welcome to Auto-Keras"
-        },
         {
             "location": "/classifier/",
             "text": "_validate\n\n\nCheck x_train's type and the shape of x_train, y_train.\n\n\nread_csv_file\n\n\nRead the cvs file and returns two seperate list containing images name and their labels\n\n\nArgs\n\n\ncsv_file_path\n: Path to the CVS file.\n\n\nReturns\n\n\nimg_file_names list containing images names and img_label list containing their respective labels.\n\n\nread_images\n\n\nReads the images from the path and return there numpy.ndarray instance\n\n\nArgs\n\n\nimg_file_names\n: List containing images names\n\n\nimages_dir_path\n: Path to directory containing images\n\n\nload_image_dataset\n\n\nLoad images from the files and labels from a csv file.\nSecond, the dataset is a set of images and the labels are in a CSV file. The CSV file should contain two columns whose names are 'File Name' and 'Label'. The file names in the first column should match the file names of the images with extensions, e.g., .jpg, .png. The path to the CSV file should be passed through the csv_file_path. The path to the directory containing all the images should be passed through image_path.\n\n\nArgs\n\n\ncsv_file_path\n: CVS file path.\n\n\nimages_path\n: Path where images exist.\n\n\nReturns\n\n\nx: Four dimensional numpy.ndarray. The channel dimension is the last dimension. y: The labels.\n\n\nImageClassifier\n\n\nThe image classifier class.\nIt is used for image classification. It searches convolutional neural network architectures for the best configuration for the dataset.\n\n\nAttributes\n\n\npath\n: A path to the directory to save the classifier.\n\n\ny_encoder\n: An instance of OneHotEncoder for y_train (array of categorical labels).\n\n\nverbose\n: A boolean value indicating the verbosity mode.\n\n\nsearcher\n: An instance of BayesianSearcher. It search different\n    neural architecture to find the best model.\n\n\nsearcher_args\n: A dictionary containing the parameters for the searcher's \ninit\n function.\n\n\ninit\n\n\nInitialize the instance.\nThe classifier will be loaded from the files in 'path' if parameter 'resume' is True. Otherwise it would create a new one.\n\n\nArgs\n\n\nverbose\n: An boolean of whether the search process will be printed to stdout.\n\n\npath\n: A string. The path to a directory, where the intermediate results are saved.\n\n\nresume\n: An boolean. If True, the classifier will continue to previous work saved in path.\n    Otherwise, the classifier will start a new search.\n\n\nfit\n\n\nFind the best neural architecture and train it.\nBased on the given dataset, the function will find the best neural architecture for it. The dataset is in numpy.ndarray format. So they training data should be passed through x_train, y_train.\n\n\nArgs\n\n\nx_train\n: An numpy.ndarray instance contains the training data.\n\n\ny_train\n: An numpy.ndarray instance contains the label of the training data.\n\n\ntime_limit\n: The time limit for the search in seconds.\n\n\npredict\n\n\nReturn predict result for the testing data.\n\n\nArgs\n\n\nx_test\n: An instance of numpy.ndarray contains the testing data.\n\n\nReturns\n\n\nAn numpy.ndarray containing the results.\n\n\nsummary\n\n\nPrint the summary of the best model.\n\n\nevaluate\n\n\nReturn the accuracy score between predict value and test_y.\n\n\nfinal_fit\n\n\nFinal training after found the best architecture.\n\n\nArgs\n\n\nx_train\n: An numpy.ndarray of training data.\n\n\ny_train\n: An numpy.ndarray of training targets.\n\n\nx_test\n: An numpy.ndarray of testing data.\n\n\ny_test\n: An numpy.ndarray of testing targets.\n\n\ntrainer_args\n: A dictionary containing the parameters of the ModelTrainer constructure.\n\n\nretrain\n: A boolean of whether reinitialize the weights of the model.\n\n\nexport_keras_model\n\n\nExport the searched model as a Keras saved model.\n\n\nArgs\n\n\npath\n: A string. The path to the file to save.\n\n\nmodel_id\n: A integer. If not provided, the function will export the best model.\n\n\nget_best_model_id\n\n\nReturns: An integer. The best model id.",

docs/sitemap.xml

@@ -2,42 +2,42 @@
 <urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
     <url>
      <loc>/</loc>
-     <lastmod>2018-07-01</lastmod>
+     <lastmod>2018-07-04</lastmod>
      <changefreq>daily</changefreq>
     </url>
     <url>
      <loc>/classifier/</loc>
-     <lastmod>2018-07-01</lastmod>
+     <lastmod>2018-07-04</lastmod>
      <changefreq>daily</changefreq>
     </url>
     <url>
      <loc>/search/</loc>
-     <lastmod>2018-07-01</lastmod>
+     <lastmod>2018-07-04</lastmod>
      <changefreq>daily</changefreq>
     </url>
     <url>
      <loc>/graph/</loc>
-     <lastmod>2018-07-01</lastmod>
+     <lastmod>2018-07-04</lastmod>
      <changefreq>daily</changefreq>
     </url>
     <url>
      <loc>/preprocessor/</loc>
-     <lastmod>2018-07-01</lastmod>
+     <lastmod>2018-07-04</lastmod>
      <changefreq>daily</changefreq>
     </url>
     <url>
      <loc>/utils/</loc>
-     <lastmod>2018-07-01</lastmod>
+     <lastmod>2018-07-04</lastmod>
      <changefreq>daily</changefreq>
     </url>
     <url>
      <loc>/generator/</loc>
-     <lastmod>2018-07-01</lastmod>
+     <lastmod>2018-07-04</lastmod>
      <changefreq>daily</changefreq>
     </url>
     <url>
      <loc>/about/</loc>
-     <lastmod>2018-07-01</lastmod>
+     <lastmod>2018-07-04</lastmod>
      <changefreq>daily</changefreq>
     </url>
 </urlset>

mkdocs/docs/index.md

@@ -1,9 +1,9 @@
-# Welcome to Auto-Keras
+<img src="https://github.com/jhfjhfj1/autokeras/blob/docs/logo.png?raw=true" alt="drawing" width="400px"/>
 
 [![Build Status](https://travis-ci.org/jhfjhfj1/autokeras.svg?branch=master)](https://travis-ci.org/jhfjhfj1/autokeras)
 [![Coverage Status](https://coveralls.io/repos/github/jhfjhfj1/autokeras/badge.svg?branch=master)](https://coveralls.io/github/jhfjhfj1/autokeras?branch=master)
 
-This is a automated machine learning (AutoML) package based on Keras. 
+This is a automated machine learning (AutoML) package based on Keras.
 It aims at automatically search for the architecture and hyperparameters for deep learning models.
 The ultimate goal for this project is for domain experts in fields other than computer science or machine learning
 to use deep learning models conveniently.
@@ -14,9 +14,9 @@ To install the package please use the commend as follows:
 
 Here is a short example for using the package.
 
-    
+
     import autokeras as ak
-    
+
     (x_train, y_train), (x_test, y_test) = mnist.load_data()
     clf = ak.ImageClassifier()
     clf.fit(x_train, y_train)