Updated Readme

andersx · andersx · commit 1ef1b29aa239 · 2017-05-13T16:11:55.000+02:00
diff --git a/README.md b/README.md
@@ -96,41 +96,47 @@ Generate ~100K atomic coulomb matrices = 0.22s
 
 Example 1: Using a coulomb matrix
 
-    from qml.kernels import laplacian_kernel
-
-    comps = ... # load a Python list of Compunds
+```python
+from qml.kernels import laplacian_kernel
 
-    for comp in comps:
-        comp.generate_coulomb_matrix()
+comps = ... # load a Python list of Compunds
 
-    training = comps[:1000]
+for comp in comps:
+    comp.generate_coulomb_matrix()
 
-    K = laplacian_kernel(training, training, sigma=100.0) 
+training = comps[:1000]
 
+K = laplacian_kernel(training, training, sigma=100.0) 
+```
 
 Example 1: Using ARAD representation 
 
-    from qml.kernels import arad_kernel
+```python
+from qml.kernels import arad_kernel
 
-    comps = ... # load a Python list of Compunds
+comps = ... # load a Python list of Compunds
 
-    for comp in comps:
-        comp.generate_coulomb_matrix()
+for comp in comps:
+    comp.generate_coulomb_matrix()
 
-    training = comps[:1000]
+training = comps[:1000]
 
-    K = arad_kernel(training, training, sigma=100.0)
+K = arad_kernel(training, training, sigma=100.0)
+```
 
 Additionally for ARAD, when the two sets of `Compound` are identical, it is possible to calculate only the upper triangle, which reduces the computational load by a factor of two.
 
-    from qml.kernels import arad_training_kernel
-    K = arad_training_kernel(training, sigma=100.0) 
-
+```python
+from qml.kernels import arad_training_kernel
+K = arad_training_kernel(training, sigma=100.0) 
+```
 
 ## 3.2) Calculate kernels using the `qml.kernels` module directly
 
 
-    from qml.kernels import laplacian_kernel
+```python
+from qml.kernels import laplacian_kernel
+```
 
 ## 3.3) Benchmarks for QM7:
 
