Polished headers and titles for DL landing page (#8)

* Build using make script * lava-dl-slayer doc fixes * autodoc working. DL landing page modified. * Added sync scripts in makefile for tutorial notebook sync, modified changes suggested by Marcus. * Updated Readme * removed redundant copy of notebooks * polished landing page * modified sync_notebook so that it will exculde builds for libraries that do not exist * Update conf.py Co-authored-by: Marcus G K Williams <[email protected]>
lava-nc · Nov 13, 2021 · 94b64f8 · 94b64f8
1 parent 2f1c84d
commit 94b64f8
Show file tree

Hide file tree

Showing 7 changed files with 98 additions and 59 deletions.
diff --git a/algorithms.rst b/algorithms.rst
@@ -4,7 +4,7 @@ Algorithms and Application Libraries
 Following are the list of algorithms and applications implemented using Lava:
 
 .. toctree::
-   :maxdepth: 4
+   :maxdepth: 3
 
    dl
    dnf 

diff --git a/build.py b/build.py
@@ -11,6 +11,7 @@
 url = "https://lava-nc.org"
 license = "BSD-3-Clause"
 
+
 @init
 def set_properties(project):
     project.set_property("dir_source_main_python", "../lava")

diff --git a/conf.py b/conf.py
@@ -1,7 +1,3 @@
-import os
-import sys
-import sphinx_rtd_theme
-
 project = "Lava"
 copyright = "2021, Intel Corporation"
 author = "Intel Corporation"
@@ -17,7 +13,6 @@
     "nbsphinx",
 ]
 
-# napoleon_google_docstring = True
 napoleon_numpy_docstring = True
 napoleon_include_init_with_doc = False
 napoleon_include_private_with_doc = False

diff --git a/dl.rst b/dl.rst
@@ -1,7 +1,10 @@
-Lava DL
-=======
+Deep Learning
+=============
 
-``lava-dl`` is a library of deep learning tools within Lava that
+Introduction
+------------
+
+Lava-DL (``lava-dl``) is a library of deep learning tools within Lava that 
 support offline training, online training and inference methods for
 various Deep Event-Based Networks.
 
@@ -69,10 +72,10 @@ Getting Started
 
 * `Dynamics and Neurons <lava-lib-dl/slayer/notebooks/neuron_dynamics/dynamics.html>`__
 
-``lava.lib.dl.slayer``
-----------------------
+SLAYER 2.0
+----------
 
-``lava.lib.dl.slayer`` is an enhanced version of
+SLAYER 2.0 (`lava.lib.dl.slayer`) is an enhanced version of
 `SLAYER <https://github.com/bamsumit/slayerPytorch>`__. Most noteworthy
 enhancements are: support for *recurrent network structures*, a wider
 variety of *neuron models* and *synaptic connections* (a complete list
@@ -81,7 +84,7 @@ of features is
 This version of SLAYER is built on top of the
 `PyTorch <https://pytorch.org/>`__ deep learning framework, similar to
 its predecessor. For smooth integration with Lava,
-``lava.lib.dl.slayer`` supports exporting trained models using the
+`lava.lib.dl.slayer` supports exporting trained models using the
 platform independent **hdf5 network exchange** format.
 
 In future versions, SLAYER will get completely integrated into Lava to
@@ -151,15 +154,15 @@ Example Code
 
    net.export_hdf5('network.net')
 
-``lava.lib.dl.bootstrap``
--------------------------
+Bootstrap
+---------
 
 In general ANN-SNN conversion methods for rate based SNN result in high latency of the network during inference. This is because the rate interpretation of a spiking neuron using ReLU acitvation unit breaks down for short inference times. As a result, the network requires many time steps per sample to achieve adequate inference results.
 
-``lava.lib.dl.bootstrap`` enables rapid training of rate based SNNs by translating them to an equivalent dynamic ANN representation which leads to SNN performance close to the equivalent ANN and low latency inference. More details `here <lava-lib-dl/bootstrap/bootstrap.html>`__. It also supports *hybrid training*
+Bootstrap (`lava.lib.dl.bootstrap`) enables rapid training of rate based SNNs by translating them to an equivalent dynamic ANN representation which leads to SNN performance close to the equivalent ANN and low latency inference. More details `here <lava-lib-dl/bootstrap/bootstrap.html>`__. It also supports *hybrid training*
 a mixed ANN-SNN network to minimize the ANN to SNN performance gap. This method is independent of the SNN model being used.
 
-It has similar API as ``lava.lib.dl.slayer`` and supports exporting
+It has similar API as `lava.lib.dl.slayer` and supports exporting
 trained models using the platform independent **hdf5 network exchange**
 format.
 
@@ -231,10 +234,10 @@ Example Code
 
    net.export_hdf5('network.net')
 
-``lava.lib.dl.netx`` 
---------------------
+Network Exchange (NetX) Library 
+-------------------------------
 
-For inference using Lava, ``lava.lib.dl.netx`` provides an
+For inference using Lava, Network Exchange Library (`lava.lib.dl.netx`) provides an
 automated API for loading SLAYER-trained models as Lava Processes, which
 can be directly run on a desired backend. ``lava.lib.dl.netx`` imports
 models saved via SLAYER using the hdf5 network exchange format. The
@@ -286,6 +289,9 @@ Example Code
 
    net.run(condition=rcnd.RunSteps(total_run_time), run_cfg=rcfg.Loihi1SimCfg())
 
+Detailed Description
+--------------------
+
 .. toctree::
    :maxdepth: 1
    :caption: Detailed description:

diff --git a/lava-lib-dl/netx/index.rst b/lava-lib-dl/netx/index.rst
@@ -1,5 +1,5 @@
-Network Exchange (Netx)
-=======================
+Lava-DL NetX
+============
 
 .. Note::
    Coming soon.
diff --git a/lava-lib-dl/netx/netx.rst b/lava-lib-dl/netx/netx.rst
@@ -1,7 +1,7 @@
-Lava-dl-netx
+Lava-DL NetX
 ============
 
-``lava.lib.dl.netx`` automates deep learning network model exchange
+Lava-DL NetX library ``lava.lib.dl.netx`` automates deep learning network model exchange
 to/from LAVA from/to other frameworks. At the moment, we support a
 simple platform independent hdf5 network description protocol. In
 furture we will extend network exchange support to other neural network

diff --git a/sync_notebook.py b/sync_notebook.py
@@ -2,21 +2,43 @@
 import glob
 from distutils.dir_util import copy_tree
 
-import lava
-import lava.lib.dl.slayer as slayer
-import lava.lib.dl.bootstrap as bootstrap
+module_dict = {}
+
+try:
+    import lava
+    module_dict.update({'lava': lava})
+except ModuleNotFoundError:
+    print(f"Failed importing {lava}. It's dependencies will be excluded.")
+
+try:
+    import lava.lib.dl.slayer as slayer
+    module_dict.update({'slayer': slayer})
+except ModuleNotFoundError:
+    print(f"Failed importing slayer. It's dependencies will be excluded.")
+
+try:
+    import lava.lib.dl.bootstrap as bootstrap
+    module_dict.update({'bootstrap': bootstrap})
+except ModuleNotFoundError:
+    print(f"Failed importing bootstrap. It's dependencies will be excluded.")
+
+try:
+    import lava.lib.optimization as optim
+    module_dict.update({'optim': optim})
+except ModuleNotFoundError:
+    print(f"Failed importing optim. It's dependencies will be excluded.")
 
 
 tutorial_list = [  # list of all notebooks to sync
     {
-        'module': lava,
+        'module': 'lava',
         'dst': 'lava/notebooks/',
         'tutorials': {
             'end_to_end': 'End to end tutorials',
         },
     },
     {
-        'module': slayer,
+        'module': 'slayer',
         'dst': 'lava-lib-dl/slayer/notebooks/',
         'tutorials': {
             'oxford': 'Oxford spike train regression',
@@ -26,12 +48,19 @@
         },
     },
     {
-        'module': bootstrap,
+        'module': 'bootstrap',
         'dst': 'lava-lib-dl/bootstrap/notebooks/',
         'tutorials': {
             'mnist': 'MNIST digit classification',
         },
     },
+    {
+        'module': 'optim',
+        'dst': 'lava/notebooks/',
+        'tutorials': {
+            'end_to_end': 'End to end tutorials',
+        },
+    },
 ]
 
 
@@ -65,40 +94,48 @@ def create_nb_rst(folder_path, rst_name, header, ignore=[]):
             continue
         rst_text += f'    {nb_name}{os.linesep}'
 
+
+    print('Creating ' + folder_path + '/' + rst_name)
     with open(folder_path + '/' + rst_name, 'wt') as f:
         f.write(rst_text)
 
 
 if __name__ == '__main__':
     for tutorials in tutorial_list:
-        module = tutorials['module']
-        module_path = module.__path__[0]
-        module_path = module_path.split('src/lava')[0]
-        if module_path[-1] != '/':  # this is temp before lava dir restructure
-            module_path += '/'
-        dst = tutorials['dst']
-        if 'ignore' in tutorials.keys():
-            ignore = tutorials['ignore']
-        else:
-            ignore = []
-        os.makedirs(dst, exist_ok=True)
-        for tutorial, header in tutorials['tutorials'].items():
-            src_path = glob.glob(
-                f'{module_path}tutorials/**/{tutorial}',
-                recursive=True,
-            )
-            dst_path = dst + tutorial
-            if len(src_path) == 1:
-                src_path = src_path[0]
-                print(f'copying from {src_path} to {dst_path}')
-                copy_tree(src_path, dst_path)
-                create_nb_rst(src_path, tutorial+'.rst', header, ignore)
+        key = tutorials['module']
+        if key in module_dict.keys():
+            module = module_dict[key]
+            module_path = module.__path__[0]
+            module_path = module_path.split('src/lava')[0]
+            if module_path[-1] != '/': 
+                # this is temp before lava dir restructure
+                module_path += '/'
+            dst = tutorials['dst']
+            if 'ignore' in tutorials.keys():
+                ignore = tutorials['ignore']
             else:
-                if len(src_path) == 0:
-                    print(f'search path: {module_path}tutorials/**/{tutorial}')
-                    raise Exception('Module not found! Check your config')
+                ignore = []
+            os.makedirs(dst, exist_ok=True)
+            for tutorial, header in tutorials['tutorials'].items():
+                src_path = glob.glob(
+                    f'{module_path}tutorials/**/{tutorial}',
+                    recursive=True,
+                )
+                dst_path = dst + tutorial
+                if len(src_path) == 1:
+                    src_path = src_path[0]
+                    print(f'copying from {src_path} to {dst_path}')
+                    copy_tree(src_path, dst_path)
+                    create_nb_rst(dst_path, tutorial+'.rst', header, ignore)
                 else:
-                    raise Exception(
-                        f'Multiple Moudles found: '
-                        f'{src_path}'
-                    )
+                    if len(src_path) == 0:
+                        print(
+                            f'search path: '
+                            f'{module_path}tutorials/**/{tutorial}'
+                        )
+                        raise Exception('Module not found! Check your config')
+                    else:
+                        raise Exception(
+                            f'Multiple Moudles found: '
+                            f'{src_path}'
+                        )