Updated godot example to use logistic regression usecase

examples/godot_logistic_regression/.gitignore

@@ -0,0 +1,2 @@
+.import
+godot_engine/godot

examples/godot_logistic_regression/custom_module/.gitignore

@@ -0,0 +1,3 @@
+vulkan-kompute
+lib
+godot

examples/godot_logistic_regression/custom_module/README.md

@@ -0,0 +1,66 @@
+# Vulkan Kompute Godot Example
+
+## Set Up Dependencies
+
+### Vulkan
+
+You will need the Vulkan SDK, in this case we use version `1.2.148.1`, which you can get at the official site https://vulkan.lunarg.com/sdk/home#windows
+
+This will have the following contents that will be required later on:
+
+* The VulkanSDK static library `vulkan-1`
+
+### Kompute
+
+We will be using v0.3.1 of Kompute, and similar to above we will need the built static library, but in this case we will build it.
+
+We can start by cloning the repository on the v0.3.1 branch:
+
+```
+git clone --branch v0.3.1 https://github.com/EthicalML/vulkan-kompute/
+```
+
+You will be able to use cmake to generate the build files for your platform.
+
+```
+cmake vulkan-kompute/. -Bvulkan-kompute/build
+```
+
+You need to make sure that the build is configured with the same flags required for godot, for example, in windows you will need:
+
+* Release build
+* Configuration type: static library
+* Runtime lib: Multi-threaded / multi-threaded debug
+
+Now you should see the library built under `build/src/Release`
+
+## Building Godot
+
+Now to build godot you will need to set up a couple of things for the Scons file to work - namely setting up the following:
+
+* Copy the `vulkan-1` library from your vulkan sdk folder to `lib/vulkan-1.lib`
+* Copy the `kompute.lib` library from the Kompute build to `lib/kompute.lib`
+* Make sure the versions above match as we provide the headers in the `include` folder - if you used different versions make sure these match as well
+
+### Clone godot repository
+
+Now we can clone the godot repository - it must be on a separate repository, so you can use the parent directory if you are on the Kompute repo.
+
+```
+cd ../../godot_engine
+
+git clone --branch 3.2.3-stable https://github.com/godotengine/godot
+
+cd godot/
+```
+
+And now we can build against our module
+
+```
+wscons -j16 custom_modules=../../custom_module/ platform=windows target=release_debug
+```
+
+Once we have built it we can now run the generated godot engine in the `bin/` folder, and we will be able to access the custom module from anywhere in the project, as well as creating new nodes from the user interface.
+
+
+

examples/godot_logistic_regression/custom_module/kompute_model_ml/KomputeModelMLNode.cpp

@@ -0,0 +1,154 @@
+/* summator.cpp */
+
+#include <vector>
+
+#include "KomputeModelMLNode.h"
+
+KomputeModelMLNode::KomputeModelMLNode() {
+    std::cout << "CALLING CONSTRUCTOR" << std::endl;
+    this->_init();
+}
+
+void KomputeModelMLNode::train(Array yArr, Array xIArr, Array xJArr) {
+
+    assert(yArr.size() == xIArr.size());
+    assert(xIArr.size() == xJArr.size());
+
+    std::vector<float> yData;
+    std::vector<float> xIData;
+    std::vector<float> xJData;
+    std::vector<float> zerosData;
+
+    for (size_t i = 0; i < yArr.size(); i++) {
+        yData.push_back(yArr[i]);
+        xIData.push_back(xIArr[i]);
+        xJData.push_back(xJArr[i]);
+        zerosData.push_back(0);
+    }
+
+    uint32_t ITERATIONS = 100;
+    float learningRate = 0.1;
+
+    std::shared_ptr<kp::Tensor> xI{ new kp::Tensor(xIData) };
+    std::shared_ptr<kp::Tensor> xJ{ new kp::Tensor(xJData) };
+
+    std::shared_ptr<kp::Tensor> y{ new kp::Tensor(yData) };
+
+    std::shared_ptr<kp::Tensor> wIn{ new kp::Tensor({ 0.001, 0.001 }) };
+    std::shared_ptr<kp::Tensor> wOutI{ new kp::Tensor(zerosData) };
+    std::shared_ptr<kp::Tensor> wOutJ{ new kp::Tensor(zerosData) };
+
+    std::shared_ptr<kp::Tensor> bIn{ new kp::Tensor({ 0 }) };
+    std::shared_ptr<kp::Tensor> bOut{ new kp::Tensor(zerosData) };
+
+    std::shared_ptr<kp::Tensor> lOut{ new kp::Tensor(zerosData) };
+
+    std::vector<std::shared_ptr<kp::Tensor>> params = { xI,  xJ,    y,
+                                                        wIn, wOutI, wOutJ,
+                                                        bIn, bOut,  lOut };
+
+    {
+        kp::Manager mgr;
+
+        if (std::shared_ptr<kp::Sequence> sq =
+              mgr.getOrCreateManagedSequence("createTensors").lock()) {
+
+            sq->begin();
+
+            sq->record<kp::OpTensorCreate>(params);
+
+            sq->end();
+            sq->eval();
+
+            // Record op algo base
+            sq->begin();
+
+            sq->record<kp::OpTensorSyncDevice>({ wIn, bIn });
+
+            sq->record<kp::OpAlgoBase<>>(
+              params, std::vector<char>(LR_SHADER.begin(), LR_SHADER.end()));
+
+            sq->record<kp::OpTensorSyncLocal>({ wOutI, wOutJ, bOut, lOut });
+
+            sq->end();
+
+            // Iterate across all expected iterations
+            for (size_t i = 0; i < ITERATIONS; i++) {
+
+                sq->eval();
+
+                for (size_t j = 0; j < bOut->size(); j++) {
+                    wIn->data()[0] -= learningRate * wOutI->data()[j];
+                    wIn->data()[1] -= learningRate * wOutJ->data()[j];
+                    bIn->data()[0] -= learningRate * bOut->data()[j];
+                }
+            }
+        }
+    }
+
+    SPDLOG_INFO("RESULT: <<<<<<<<<<<<<<<<<<<");
+    SPDLOG_INFO(wIn->data()[0]);
+    SPDLOG_INFO(wIn->data()[1]);
+    SPDLOG_INFO(bIn->data()[0]);
+
+    this->mWeights = kp::Tensor(wIn->data());
+    this->mBias = kp::Tensor(bIn->data());
+}
+
+Array KomputeModelMLNode::predict(Array xI, Array xJ) {
+    assert(xI.size() == xJ.size());
+
+    Array retArray;
+
+    // We run the inference in the CPU for simplicity
+    // BUt you can also implement the inference on GPU 
+    // GPU implementation would speed up minibatching
+    for (size_t i = 0; i < xI.size(); i++) {
+        float xIVal = xI[i];
+        float xJVal = xJ[i];
+        float result = (xIVal * this->mWeights.data()[0]
+                + xJVal * this->mWeights.data()[1]
+                + this->mBias.data()[0]);
+
+        // Instead of using sigmoid we'll just return full numbers
+        Variant var = result > 0 ? 1 : 0;
+        retArray.push_back(var);
+    }
+
+    return retArray;
+}
+
+Array KomputeModelMLNode::get_params() {
+    Array retArray;
+
+    SPDLOG_INFO(this->mWeights.size() + this->mBias.size());
+
+    if(this->mWeights.size() + this->mBias.size() == 0) {
+        return retArray;
+    }
+
+    retArray.push_back(this->mWeights.data()[0]);
+    retArray.push_back(this->mWeights.data()[1]);
+    retArray.push_back(this->mBias.data()[0]);
+    retArray.push_back(99.0);
+
+    return retArray;
+}
+
+void KomputeModelMLNode::_init() {
+    std::cout << "CALLING INIT" << std::endl;
+}
+
+void KomputeModelMLNode::_process(float delta) {
+
+}
+
+void KomputeModelMLNode::_bind_methods() {
+    ClassDB::bind_method(D_METHOD("_process", "delta"), &KomputeModelMLNode::_process);
+    ClassDB::bind_method(D_METHOD("_init"), &KomputeModelMLNode::_init);
+
+    ClassDB::bind_method(D_METHOD("train", "yArr", "xIArr", "xJArr"), &KomputeModelMLNode::train);
+    ClassDB::bind_method(D_METHOD("predict", "xI", "xJ"), &KomputeModelMLNode::predict);
+    ClassDB::bind_method(D_METHOD("get_params"), &KomputeModelMLNode::get_params);
+}
+

examples/godot_logistic_regression/custom_module/kompute_model_ml/KomputeModelMLNode.h

@@ -0,0 +1,88 @@
+#pragma once
+
+#include <memory>
+
+#include "kompute/Kompute.hpp"
+
+#include "scene/main/node.h"
+
+class KomputeModelMLNode : public Node {
+    GDCLASS(KomputeModelMLNode, Node);
+
+public:
+    KomputeModelMLNode();
+
+    void train(Array y, Array xI, Array xJ);
+
+    Array predict(Array xI, Array xJ);
+
+    Array get_params();
+
+    void _process(float delta);
+    void _init();
+
+protected:
+    static void _bind_methods();
+
+private:
+    kp::Tensor mWeights;
+    kp::Tensor mBias;
+};
+
+static std::string LR_SHADER = R"(
+#version 450
+
+layout (constant_id = 0) const uint M = 0;
+
+layout (local_size_x = 1) in;
+
+layout(set = 0, binding = 0) buffer bxi { float xi[]; };
+layout(set = 0, binding = 1) buffer bxj { float xj[]; };
+layout(set = 0, binding = 2) buffer by { float y[]; };
+layout(set = 0, binding = 3) buffer bwin { float win[]; };
+layout(set = 0, binding = 4) buffer bwouti { float wouti[]; };
+layout(set = 0, binding = 5) buffer bwoutj { float woutj[]; };
+layout(set = 0, binding = 6) buffer bbin { float bin[]; };
+layout(set = 0, binding = 7) buffer bbout { float bout[]; };
+layout(set = 0, binding = 8) buffer blout { float lout[]; };
+
+float m = float(M);
+
+float sigmoid(float z) {
+    return 1.0 / (1.0 + exp(-z));
+}
+
+float inference(vec2 x, vec2 w, float b) {
+    // Compute the linear mapping function
+    float z = dot(w, x) + b;
+    // Calculate the y-hat with sigmoid 
+    float yHat = sigmoid(z);
+    return yHat;
+}
+
+float calculateLoss(float yHat, float y) {
+    return -(y * log(yHat)  +  (1.0 - y) * log(1.0 - yHat));
+}
+
+void main() {
+    uint idx = gl_GlobalInvocationID.x;
+
+    vec2 wCurr = vec2(win[0], win[1]);
+    float bCurr = bin[0];
+
+    vec2 xCurr = vec2(xi[idx], xj[idx]);
+    float yCurr = y[idx];
+
+    float yHat = inference(xCurr, wCurr, bCurr);
+
+    float dZ = yHat - yCurr;
+    vec2 dW = (1. / m) * xCurr * dZ;
+    float dB = (1. / m) * dZ;
+    wouti[idx] = dW.x;
+    woutj[idx] = dW.y;
+    bout[idx] = dB;
+
+    lout[idx] = calculateLoss(yHat, yCurr);
+}
+)";
+

examples/godot_logistic_regression/custom_module/kompute_model_ml/SCsub

@@ -0,0 +1,17 @@
+import os
+
+Import('env')
+
+dir_path = os.getcwd()
+
+# Kompute & Vulkan header files
+env.Append(CPPPATH = ['include/'])
+
+env.add_source_files(env.modules_sources, "*.cpp")
+
+# Kompute & Vulkan libraries
+env.Append(LIBS=[
+        File(dir_path +'/lib/kompute.lib'),
+        File(dir_path +'/lib/vulkan-1.lib'),
+    ])
+

examples/godot_logistic_regression/custom_module/kompute_model_ml/config.py

@@ -0,0 +1,5 @@
+def can_build(env, platform):
+    return True
+
+def configure(env):
+    pass