neworderofjamie

#include <iostream>
#include <memory>
#include <string>
#include <vector>


// Forward declarations
class NoAnnotation;

template<typename A = NoAnnotation>

neworderofjamie

class NumericTypeValue
{
public:
    NumericTypeValue(float value) : m_Value(double{value}){}
    NumericTypeValue(double value) : m_Value(value){}
    NumericTypeValue(int value): m_Value(int64_t{value}){}
    NumericTypeValue(unsigned int value): m_Value(uint64_t{value}){}
    NumericTypeValue(uint64_t value): m_Value(value){}
    NumericTypeValue(int64_t value): m_Value(value){}

neworderofjamie

import cv2
import numpy as np
import os
import sys
import csv

def get_box_centre(box):
    return (int(box[0] + (box[2] / 2)), int(box[1] + (box[3] / 2)))

def draw_box(frame, box, colour):

neworderofjamie

import cv2
import numpy as np
import os
import sys
import csv

def draw_box(frame, box, colour):
    (x, y, w, h) = (int(v) for v in box)
    cv2.rectangle(frame, (x, y), (x + w, y + h), colour, 2)
    

neworderofjamie

neworderofjamie

# Create sequential model
model = SequentialModel()
with model:
    input = InputLayer(IntegrateFireInput(v_thresh=5.0), 784)
    Layer(Dense(weight=weights[0]), IntegrateFire(v_thresh=5.0))
    Layer(Dense(weight=weights[1]), IntegrateFire(v_thresh=5.0, output=SpikeCount()))

neworderofjamie

import numpy as np
from os import path

from ml_genn import InputLayer, Layer, SequentialModel
from ml_genn.compilers import Compiler
from ml_genn.neurons import IntegrateFire, IntegrateFireInput
from ml_genn.connectivity import Dense

# Load weights
weights = []

neworderofjamie

import numpy as np
import matplotlib.pyplot as plt

spikes = np.loadtxt("spikes.csv", delimiter=",")
voltages = np.loadtxt("voltages.csv", delimiter=",")

fig, axes = plt.subplots(5, sharex=True)

for i in range(4):
    axes[i].plot(voltages[:,0], voltages[:,i+1])

neworderofjamie

// C++ includes
#include <algorithm>
#include <fstream>

// ------------------------------------------------------------------------
// Helper macro for error-checking CUDA calls
#define CHECK_CUDA_ERRORS(call) {\
    cudaError_t error = call;\
    if (error != cudaSuccess) {\
        throw std::runtime_error(__FILE__": " + std::to_string(__LINE__) + ": cuda error " + std::to_string(error) + ": " + cudaGetErrorString(error));\

neworderofjamie

import os, pathlib
import tensorflow as tf
from tensorflow.keras import (models, layers, datasets, callbacks, optimizers,
                              initializers, regularizers)
from tensorflow.keras.utils import CustomObjectScope
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from glob import glob
import numpy as np

# Learning rate schedule