diff --git a/README.md b/README.md
index 4a91ebc..37ed7aa 100644
--- a/README.md
+++ b/README.md
@@ -28,7 +28,7 @@ Assuming you have [Scikit-Learn](https://scikit-learn.org/) already installed, y
 ```python
 from sklearn import svm
 from stringkernels.kernels import string_kernel
-model = svm.SVC(kernel=string_kernel)
+model = svm.SVC(kernel=string_kernel())
 ```
 
 and the polynomial string kernel,
@@ -36,9 +36,9 @@ and the polynomial string kernel,
 ```python
 from sklearn import svm
 from stringkernels.kernels import polynomial_string_kernel
-model = svm.SVC(kernel=polynomial_string_kernel)
+model = svm.SVC(kernel=polynomial_string_kernel())
 ```
 
-See the notebook [example.ipynb](https://github.com/weekend37/string-kernels/blob/master/example.ipynb) for further demonstration of usage.
+For morer information read the [docs](https://github.com/weekend37/string-kernels/blob/master/doc/docs.md) or take a look at the notebook [example.ipynb](https://github.com/weekend37/string-kernels/blob/master/example.ipynb) for further demonstration of usage.
 
 If you end up using this in your research we kindly ask you to cite us! :)
diff --git a/doc/fig/docs.md b/doc/fig/docs.md
new file mode 100644
index 0000000..4cb02e4
--- /dev/null
+++ b/doc/fig/docs.md
@@ -0,0 +1,54 @@
+# Documentation
+
+## kernels.string_kernel
+
+**Wrapper for a singly vectorized linear time string kernel implentation for data matrices X and Y**
+```python
+    Parameters
+        - normalzie : bool, default=True
+            indicates if the kernel output should be normalized s.t. max(K) <= 1
+        - n_jobs : int, default=None
+            how many CPUs to distribute the process over. If None, use maximum available CPUs.
+    
+    Returns
+        - string_kernel_func : function
+            function that takes in two data matrices X and Y as arguments
+            (np.ndarray's of shapes (NX,MX) and (NY, MY) where N_ is the number of samples and M_ is sequence length)
+            and returns the string kernel value between product of all samples in X and Y (int, float depending on normalization)
+```
+
+**Example**
+
+```python
+from sklearn import svm
+from stringkernels.kernels import string_kernel
+model = svm.SVC(kernel=string_kernel(n_jobs=32))
+```
+
+## kernels.polynomial_string_kernel
+
+**Wrapper for a linear time polynomial string kernel distance implentation for two data matrices X and Y for a monomial with exponent p to run across n_jobs different CPUs.**
+```python
+    Parameters
+        - p: float or int, default = 1.2
+            exponent of the monomial which will be used
+        - normalzie : bool, default=True
+            indicates if the kernel output should be normalized s.t. max(K) <= 1
+        - n_jobs : int, default=None
+            how many CPUs to distribute the process over. If None, use maximum available CPUs.
+    
+    Returns
+        - polynomial_string_kernel_func : function
+            function that takes in two data matrices X and Y as arguments
+            (np.ndarray's of shapes (NX,MX) and (NY, MY) where N_ is the number of samples and M_ is sequence length)
+            and returns the polynomial string kernel value between product of all samples in X and Y (float)
+
+```
+
+**Example**
+
+```python
+from sklearn import svm
+from stringkernels.kernels import polynomial_string_kernel
+model = svm.SVC(kernel=polynomial_string_kernel(p=1.1))
+```
\ No newline at end of file