Fix linear regression (#6432)

* Fix Linear Regression Tutorial

* Small fix

docs/mxdoc.py

@@ -239,7 +239,7 @@ def _get_src_download_btn(out_prefix, langs, lines):
         for f in [ipynb, src]:
             f = f.split('/')[-1]
             btn += '<button type="button" class="btn btn-default download" '
-            btn += 'onclick="window.location=\'%s\'"><span class="glyphicon glyphicon-download-alt"></span> %s </a></button>\n' % (f, f)
+            btn += 'onclick="window.location=\'%s\'"><span class="glyphicon glyphicon-download-alt"></span> %s </button>\n' % (f, f)
     btn += '</div>\n'
     return btn
 

docs/tutorials/python/linear-regression.md

@@ -18,8 +18,8 @@ how to encode a dataset into an iterator that MXNet can use. The data used in th
 
 ```python
 #Training data
-train_data = np.array([[1,2],[3,4],[5,6],[3,2],[7,1],[6,9]])
-train_label = np.array([5,11,17,7,9,24])
+train_data = np.random.uniform(0, 1, [100, 2])
+train_label = np.array([train_data[i][0] + 2 * train_data[i][1] for i in range(100)])
 batch_size = 1
 
 #Evaluation Data
@@ -140,7 +140,7 @@ parameters of the model to fit the training data. This is accomplished using the
 
 ```python
 model.fit(train_iter, eval_iter,
-            optimizer_params={'learning_rate':0.01, 'momentum': 0.9},
+            optimizer_params={'learning_rate':0.005, 'momentum': 0.9},
             num_epoch=1000,
             batch_end_callback = mx.callback.Speedometer(batch_size, 2))
 ```