Minor edits

Author: stephenchouca

sdh-docs/index.html

@@ -219,5 +219,5 @@ <h1 id="system-requirements">System Requirements</h1>
 
 <!--
 MkDocs version : 0.17.2
-Build Date UTC : 2019-05-30 19:59:15
+Build Date UTC : 2019-05-30 23:10:47
 -->

sdh-docs/pycomputations/index.html

@@ -228,8 +228,10 @@ <h2 id="expressing-reductions">Expressing Reductions</h2>
 </p>
 <p>and demonstrates how matrix-vector multiplication can be expressed in index
 notation.  Both forms are supported by TACO:</p>
-<pre class="highlight"><code class="language-python">y[i] = A[i,j] * x[j]
-y[i] = sum(j, A[i,j] * x[j])</code></pre>
+<pre class="highlight"><code class="language-python">i, j = pytaco.get_index_vars(2)
+
+y[i] = A[i,j] * x[j]
+y[i] = pytaco.sum(j, A[i,j] * x[j])</code></pre>
 
 <p>Reductions that are not explicitly expressed are assumed to be over the
 smallest subexpression that captures all uses of the corresponding reduction
@@ -260,6 +262,7 @@ <h1 id="expressing-broadcasts">Expressing Broadcasts</h1>
 <pre class="highlight"><code class="language-python">A = pt.tensor([3,3])
 B = pt.tensor([3,3])
 C = pt.tensor([3,1])
+i, j = pt.get_index_vars(2)
 
 A[i, j] =  B[i, j] + C[i, j]  # ERROR!!</code></pre>
 
@@ -271,6 +274,7 @@ <h1 id="expressing-transposes">Expressing Transposes</h1>
 <pre class="highlight"><code class="language-python">A = pt.tensor([3,3], pt.format([dense, dense]))
 B = pt.tensor([3,3], pt.format([dense, dense]))
 C = pt.tensor([3,3], pt.format([dense, dense]))
+i, j = pt.get_index_vars(2)
 
 A[i,j] = B[i,j] + C[j,i]</code></pre>
 
@@ -286,15 +290,17 @@ <h1 id="expressing-transposes">Expressing Transposes</h1>
 <pre class="highlight"><code class="language-python">A = pt.tensor([3,3], pt.format([dense, compressed]))
 B = pt.tensor([3,3], pt.format([dense, compressed]))
 C = pt.tensor([3,3], pt.format([dense, compressed]))
+i, j = pt.get_index_vars(2)
 
-A[i,j] = B[i,j] + C[j,i]</code></pre>
+A[i,j] = B[i,j] + C[j,i]  # ERROR!!</code></pre>
 
 <p>As an alternative, you can first explicitly transpose <code>C</code> by invoking its
 <code>transpose</code> method, storing the result in a temporary, and then perform the
 addition with the already-transposed temporary:</p>
 <pre class="highlight"><code class="language-python">A = pt.tensor([3,3], pt.format([dense, compressed]))
 B = pt.tensor([3,3], pt.format([dense, compressed]))
 C = pt.tensor([3,3], pt.format([dense, compressed]))
+i, j = pt.get_index_vars(2)
 
 Ct = C.transpose([1, 0])  # Ct is also stored in the CSR format
 A[i,j] = B[i,j] + Ct[i,j]</code></pre>
@@ -306,15 +312,17 @@ <h1 id="expressing-transposes">Expressing Transposes</h1>
 <pre class="highlight"><code class="language-python">A = pt.tensor([3,3], pt.format([dense, compressed]))
 B = pt.tensor([3,3], pt.format([dense, compressed]))
 C = pt.tensor([3,3], pt.format([dense, compressed], [1, 0]))
+i, j = pt.get_index_vars(2)
 
-A[i,j] = B[i,j] + C[i,j]</code></pre>
+A[i,j] = B[i,j] + C[i,j]  # ERROR!!</code></pre>
 
 <p>We can again perform the same computation by invoking <code>transpose</code>, this time to
 repack <code>C</code> into the same CSR format as <code>A</code> and <code>B</code> before computing the 
 addition:</p>
 <pre class="highlight"><code class="language-python">A = pt.tensor([3,3], pt.format([dense, compressed]))
 B = pt.tensor([3,3], pt.format([dense, compressed]))
 C = pt.tensor([3,3], pt.format([dense, compressed], [1, 0]))
+i, j = pt.get_index_vars(2)
 
 Cp = C.transpose([0, 1], pt.format([dense, compressed]))  # Store a copy of C in the CSR format
 A[i,j] = B[i,j] + Cp[i,j]</code></pre>

sdh-docs/pytensors/index.html

@@ -216,7 +216,7 @@ <h1 id="defining-tensor-formats">Defining Tensor Formats</h1>
 <code>pytaco.format</code> object.  This requires specifying whether each tensor dimension
 is dense or sparse as well as (optionally) the order in which dimensions should
 be stored.  TACO also predefines some common tensor formats (including 
-<code>pt.csr</code>, <code>pt.csc</code> and <code>pt.csf</code>) that you can use out of the box.</p>
+<code>pt.csr</code> and <code>pt.csc</code>) that you can use out of the box.</p>
 <h1 id="initializing-tensors">Initializing Tensors</h1>
 <p>Tensors can be made by using python indexing syntax. For example, one may write
 the following: You can initialize a tensor by calling its <code>insert</code> method to
@@ -247,8 +247,8 @@ <h1 id="initializing-tensors">Initializing Tensors</h1>
 <pre class="highlight"><code class="language-python">A.pack()</code></pre>
 
 <p>You can then iterate over the nonzero elements of the tensor as follows:</p>
-<pre class="highlight"><code class="language-python">for elem in A:
-  print(elem)</code></pre>
+<pre class="highlight"><code class="language-python">for coordinates, val in A:
+  print(val)</code></pre>
 
 <h1 id="file-io">File I/O</h1>
 <p>Rather than manually constructing a tensor, you can load tensors directly from