ogmacorp / aogmaneo Goto Github PK

View Code? Open in Web Editor NEW

51.0 51.0 8.0 5.4 MB

Arduino and Desktop compatible fast online/incremental machine learning system.

Home Page: https://ogma.ai/

License: Other

C++ 97.98% CMake 2.02%

aogmaneo's People

Contributors

Stargazers

Watchers

Forkers

fergalbyrne me-datastudio heavy02011 lgv2018 qqq-tech iacore jacobeverist

aogmaneo's Issues

Explanation size of weights array Encoder

Dear AOgmaNeo team,

I'm going through the code in an attempt to understand how SPH is working and during that I saw the size of the weight's matrix/array is set to $\text{number of hidden cells} \times area \times \text{number of cells in column}$, (my translation of encoder.cppL245). I don't understand why the multiplication with the number of cells in a column is required.

code questions

Here, t is never history_size, and t2 starts at t-1. As a result, history_samples[history_size-1] is never touched here. Is this intentional?

AOgmaNeo/source/aogmaneo/actor.cpp

Lines 411 to 421 in 9e2e3c8

 int t = rand() % (history_size - params.min_steps) + params.min_steps; 

 // compute (partial) values, rest is completed in the kernel 

 float r = 0.0f; 

 float d = 1.0f; 

 for (int t2 = t - 1; t2 >= 0; t2--) { 

 r += history_samples[t2].reward * d; 

 d *= params.discount; 

 }

code does not run

Hello Eric,

I made a small program with AOgmaNeo. However, it doesn't work for same reason. The code was imitated from this cartpole python example.

How to reproduce

git clone https://github.com/iacore/AOgmaNeo --branch master
cmake -B build && cmake --build build
build/test1

the program should crash at hierarchy.cpp:214

layer_input_cis[index] = histories[l][i][t];

Is there a test suite?

the contribution text file mentioned it

ogmacorp / aogmaneo Goto Github PK

aogmaneo's People

Contributors

Stargazers

Watchers

Forkers

aogmaneo's Issues

Explanation size of weights array Encoder

code questions

code does not run

How to reproduce

Is there a test suite?

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent

	int t = rand() % (history_size - params.min_steps) + params.min_steps;

	// compute (partial) values, rest is completed in the kernel
	float r = 0.0f;
	float d = 1.0f;

	for (int t2 = t - 1; t2 >= 0; t2--) {
	r += history_samples[t2].reward * d;

	d *= params.discount;
	}