Virtual Lab: Neural Network Architectures

Authenticated as Guest

UUID: mit-neuro

EXPLORER

📄 main.py

📄 model_arch.py

📁 datasets

Interactive Editor

# Train large-scale brain-like network
import numpy as np
import matplotlib.pyplot as plt
import io
import base64

class SpikingNN:
    def __init__(self, size):
        self.weights = np.random.randn(size, size)
        self.membrane_potential = np.zeros(size)

def forward(self, spikes):
        self.membrane_potential += np.dot(self.weights, spikes)
        out_spikes = self.membrane_potential > 1.0
        self.membrane_potential[out_spikes] = 0.0
        return out_spikes

print("Initializing Spiking Neural Network...")
snn = SpikingNN(100)

spike_history = []
for _ in range(50):
    input_spikes = np.random.rand(100) > 0.9
    output = snn.forward(input_spikes)
    spike_history.append(np.sum(output))

plt.style.use("dark_background")
plt.figure(figsize=(6, 3))
plt.plot(spike_history, color="#10b981")
plt.title("SNN Spike Activity Over Time")
plt.xlabel("Time Step")
plt.ylabel("Active Neurons")

buf = io.BytesIO()
plt.savefig(buf, format="png", bbox_inches="tight", transparent=True)
buf.seek(0)
render_image(base64.b64encode(buf.read()).decode("utf-8"))

print("SNN Simulation Complete.")

root@phd-lab-vbox:~# python main.py

Waiting for execution...

root@phd-lab-vbox:~#