Merge pull request #11 from Saran-nns/dev

prerelease 0.3.1
Saran-nns · Nov 1, 2020 · 46af134 · 46af134
2 parents d4d24c6 + 0dd2150
commit 46af134
Show file tree

Hide file tree

Showing 6 changed files with 283 additions and 168 deletions.
diff --git a/.github/labeler.yml b/.github/labeler.yml
@@ -0,0 +1,13 @@
+# This workflow will triage pull requests and apply a label based on the
+# paths that are modified in the pull request.
+name: "Pull Request Labeler"
+on:
+- pull_request_target
+
+jobs:
+  triage:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/labeler@main
+      with:
+        repo-token: "${{ secrets.GITHUB_TOKEN }}"
diff --git a/.github/workflows/labeler.yml b/.github/workflows/labeler.yml
@@ -0,0 +1,13 @@
+# This workflow will triage pull requests and apply a label based on the
+# paths that are modified in the pull request.
+name: "Pull Request Labeler"
+on:
+- pull_request_target
+
+jobs:
+  triage:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/labeler@main
+      with:
+        repo-token: "${{ secrets.GITHUB_TOKEN }}"
diff --git a/README.md b/README.md
@@ -2,7 +2,7 @@
 
 SORN is a class of neuro-inspired artificial network build based on plasticity mechanisms in biological brain and mimic neocortical circuits ability of learning and adaptation through neuroplasticity mechanisms.
 
-For ease of maintanance, example use cases and the API(under developement) are moved to https://github.com/Saran-nns/PySORN_0.1 
+The network is developed as part of my Master thesis at Universität Osnabrück, Germany. For the ease of maintainance, the notebooks, use cases and the API(under developement) are moved to https://github.com/Saran-nns/PySORN_0.1 
 
 [![Build Status](https://travis-ci.org/Saran-nns/sorn.svg?branch=master)](https://travis-ci.org/Saran-nns/sorn)
 [![codecov](https://codecov.io/gh/Saran-nns/sorn/branch/master/graph/badge.svg)](https://codecov.io/gh/Saran-nns/sorn)
@@ -13,14 +13,9 @@ For ease of maintanance, example use cases and the API(under developement) are m
 [![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
 
 <h4 align="Left">SORN Reservoir and the evolution of synaptic efficacies</h4>
-<a href="url"><img src="https://raw.githubusercontent.com/Saran-nns/PySORN_0.1/master/v0.1.0/doc/images/SORN1.png" height="320" width="430"></a> <a href="url"><img src="https://raw.githubusercontent.com/Saran-nns/PySORN_0.1/master/v0.1.0/doc/images/weights.png" height="375" width="425" ></a>	
+<a href="url"><img src="https://raw.githubusercontent.com/Saran-nns/PySORN_0.1/master/v0.1.0/doc/images/SORN1.png" height="320" width="430"></a> <a href="url"><img src="https://raw.githubusercontent.com/Saran-nns/PySORN_0.1/master/v0.1.0/doc/images/weights.png" height="375" width="425" ></a> <a href="url"><img src="https://raw.githubusercontent.com/Saran-nns/PySORN_0.1/master/v0.1.0/doc/images/networkxx.jpg" height="375" width="425" ></a>
 
-<h4 align="center">Neural Connectome</h4> 
-<p align="center">
-<a href="url"><img src="https://github.com/Saran-nns/PySORN_0.1/blob/master/v0.1.0/doc/images/neuralcorrelationall.png" height="450" width="450" ></a>
-</p>
-
-#### To install the latest release:
+### To install the latest release:
 
 ```python
 pip install sorn
@@ -32,17 +27,16 @@ The library is still in alpha stage, so you may also want to install the latest
 pip install git+https://github.com/Saran-nns/sorn
 ```
 
-#### Dependencies
+### Dependencies
 SORN supports Python 3.5+ ONLY. For older Python versions please use the official Python client
 
 
-#### Usage:
-
-##### Update Network configurations
-
-Navigate to home/conda/envs/ENVNAME/Lib/site-packages/sorn
+### Usage:
 
-or if you are unsure about the directory of sorn
+#### Update Network configurations
+There are two ways to update/configure the network parameters,
+1. Navigate to home/conda/envs/ENVNAME/Lib/site-packages/sorn
+ ```or``` if you are unsure about the directory of ```sorn```
 
 Run
 
@@ -53,53 +47,61 @@ sorn.__file__
 ```
 to find the location of the sorn package
 
-Then, update/edit the configuration.ini
+Then, update/edit arguments in ```configuration.ini```
 
-
-##### Plasticity Phase
+2. Pass the arguments with valid names (listed below). This will override the default values at ```configuration.ini```
+. The allowed ```kwargs``` are,
+```Python
+kwargs_ = ['ne', 'nu', 'network_type_ee', 'network_type_ei', 'network_type_ie', 'lambda_ee','lambda_ei', 'lambda_ie', 'eta_stdp','eta_inhib', 'eta_ip', 'te_max', 'ti_max', 'ti_min', 'te_min', 'mu_ip','sigma_ip']
+```
+#### Simulation: Plasticity Phase
+The default ```ne, nu``` values are overriden by passing them while calling the ```simulate_sorn``` method.
 
 ```Python
 # Import 
-from sorn.sorn import RunSorn
+from sorn import Simulator
 
 # Sample input 
-inputs = [0.]
+num_features = 10
+time_steps = 200
+inputs = np.random.rand(num_features,timesteps)
 
 # To simulate the network; 
-matrices_dict, Exc_activity, Inh_activity, Rec_activity, num_active_connections = RunSorn(phase='Plasticity', matrices=None,
-                                                                          time_steps=100).run_sorn(inputs)
+matrices_dict, Exc_activity, Inh_activity, Rec_activity, num_active_connections = Simulator.simulate_sorn(inputs = inputs, phase='Plasticity', matrices=None, noise = True, time_steps=time_steps, ne = 200, nu=num_features)
 
 # To resume the simulation, load the matrices_dict from previous simulation;
-matrices_dict, Exc_activity, Inh_activity, Rec_activity, num_active_connections = RunSorn(phase='Plasticity', matrices=matrices_dict,
-                                                                          time_steps=100).run_sorn(inputs)
+matrices_dict, Exc_activity, Inh_activity, Rec_activity, num_active_connections = Simulator.simulate_sorn(inputs = inputs, phase='Plasticity', matrices=matrices_dict, noise= True, time_steps=time_steps,ne = 200, nu=num_features)
 ```
 
-##### Training phase:
+#### Training phase:
 
 ```Python
-matrices_dict, Exc_activity, Inh_activity, Rec_activity, num_active_connections = RunSorn(phase='Training', matrices=matrices_dict,
-                                                                          time_steps=100).run_sorn(inputs)
+inputs = np.random.rand(num_features,1) 
+
+# SORN network is frozen during training phase
+matrices_dict, Exc_activity, Inh_activity, Rec_activity, num_active_connections = Trainer.train_sorn(inputs = inputs, phase='Training', matrices=matrices_dict,nu=num_features, time_steps=1)
 ```
 
-#### Network Output Descriptions:
-    matrices_dict  - Dictionary of connection weights ('Wee','Wei','Wie') , Excitatory network activity ('X'), Inhibitory network activities('Y'), Threshold values ('Te','Ti')
+### Network Output Descriptions:
+  ```matrices_dict```  - Dictionary of connection weights ('Wee','Wei','Wie') , Excitatory network activity ('X'), Inhibitory network activities('Y'), Threshold values ('Te','Ti')
+
+  ```Exc_activity``` - Collection of Excitatory network activity of entire simulation period
 
-    Exc_activity - Collection of Excitatory network activity of entire simulation period
+  ```Inh_activity``` - Collection of Inhibitory network activity of entire simulation period
 
-    Inh_activitsy - Collection of Inhibitory network activity of entire simulation period
+  ```Rec_activity``` - Collection of Recurrent network activity of entire simulation period
 
-    Rec_activity - Collection of Recurrent network activity of entire simulation period
+  ```num_active_connections``` - List of number of active connections in the Excitatory pool at each time step 
 
-    num_active_connections - List of number of active connections in the Excitatory pool at each time step 
 
+### Sample use with OpenAI gym :
+#### Cartpole balance problem
+Without changing the default network parameters. 
 
-#### Sample use with OpenAI gym :
-##### Cartpole balance problem
 ```python
 # Imports
 
-import utils.InitHelper as initializer
-from sorn.sorn import Sorn, Plasticity, TrainSorn, TrainSornPlasticity
+from sorn.sorn import Simulator, Trainer
 import gym
 
 # Load the simulated network matrices
@@ -108,7 +110,6 @@ import gym
 with open('simulation_matrices.pkl','rb') as f:  
     sim_matrices,excit_states,inhib_states,recur_states,num_reservoir_conn = pickle.load(f)
 
-
 # Training parameters
 
 NUM_EPISODES = 2e6
@@ -122,21 +123,15 @@ for EPISODE in range(NUM_EPISODES):
     state = env.reset()[None,:]
 
     # Play the episode
-
     while True:
-
       if EPISODE < NUM_PLASTICITY_EPISODE:
 
         # Plasticity phase
-        sim_matrices,excit_states,inhib_states,recur_states,num_reservoir_conn = TrainSornPlasticity.train_sorn(phase = 'Plasticity',
-                                                                                                            matrices = sim_matrices,
-                                                                                                            inputs = state)
+        sim_matrices,excit_states,inhib_states,recur_states,num_reservoir_conn = Simulator.simulate_sorn(inputs = state, phase ='Plasticity', matrices = sim_matrices, noise=False)
 
       else:
         # Training phase with frozen reservoir connectivity
-        sim_matrices,excit_states,inhib_states,recur_states,num_reservoir_conn = TrainSorn.train_sorn(phase = 'Training',
-                                                                                                            matrices = sim_matrices,
-                                                                                                            inputs = state)
+        sim_matrices,excit_states,inhib_states,recur_states,num_reservoir_conn = Trainer.train_sorn(inputs = state, phase = 'Training', matrices = sim_matrices, noise= False)
 
       # Feed excit_states as input states to your RL algorithm, below goes for simple policy gradient algorithm
       # Sample policy w.r.t excitatory states and take action in the environment
@@ -149,8 +144,7 @@ for EPISODE in range(NUM_EPISODES):
         break
 ```
 
-
-#### Sample Plotting functions 
+### Sample Plotting functions 
 
 ```Python
 from sorn.utils import Plotter
@@ -164,7 +158,7 @@ Plotter.scatter_plot(spike_train = np.asarray(Exc_activity), savefig=False)
 Plotter.raster_plot(spike_train = np.asarray(Exc_activity), savefig=False)
 ```
 
-#### Sample Statistical analysis functions
+### Sample Statistical analysis functions
 
 ```Python
 from sorn.utils import Statistics
@@ -175,7 +169,7 @@ Statistics.autocorr(firing_rates = [1,1,5,6,3,7],t= 2)
 Statistics.fanofactor(spike_train= np.asarray(Exc_activity),neuron = 10,window_size = 10)
 ```
 
-#### The network is inspired by folowing articles:
+### The network is inspired by following articles:
 
 Lazar, A. (2009). SORN: a Self-organizing Recurrent Neural Network. Frontiers in Computational Neuroscience, 3. https://doi.org/10.3389/neuro.10.023.2009
 

diff --git a/sorn/__init__.py b/sorn/__init__.py
@@ -0,0 +1,7 @@
+from . import sorn
+import logging
+
+__author__ = "Saranraj Nambusubramaniyan"
+__version__ = "0.2.10"
+
+logging.basicConfig(level=logging.INFO)