Witllm/binary/mnist.py

import os

os.environ["CUDA_LAUNCH_BLOCKING"] = "1"


import torch
import torch.nn as nn
import torch.optim as optim
import torchvision
from torchvision import transforms
from torch.utils.data import DataLoader
import math
import torch.nn.functional as F
import numpy as np

torch.manual_seed(1234)
np.random.seed(1234)
torch.cuda.manual_seed_all(1234)


device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
# device = "cpu"
# torch.set_num_threads(16)
print(f"Using device: {device}")

transform = transforms.Compose(
    [transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))]  # MNIST数据集的均值和标准差
)

train_dataset = torchvision.datasets.MNIST(root="./data", train=True, download=True, transform=transform)
test_dataset = torchvision.datasets.MNIST(root="./data", train=False, download=True, transform=transform)

train_loader = DataLoader(train_dataset, batch_size=16, shuffle=True)
test_loader = DataLoader(test_dataset, batch_size=1024, shuffle=False)


def to_binary_tensor(input_tensor, bits):
    int_tensor = torch.round(input_tensor).clamp(0, 2**bits - 1).to(torch.int64)
    shifts = torch.arange(bits - 1, -1, -1, device=int_tensor.device)
    binary_bits = (int_tensor.unsqueeze(-1) >> shifts) & 1
    return binary_bits


class Lut(torch.autograd.Function):
    # input [batch,count,bits]
    # weight [count,2**bits]
    @staticmethod
    def forward(ctx, input, weight):
        batch = input.shape[0]
        count = input.shape[1]
        bits = input.shape[2]
        assert int(math.log2(weight.shape[-1])) == bits

        index = 2 ** torch.arange(bits - 1, -1, -1, device=input.device)
        x = (input > 0).long()
        x = x * index
        ind = x.sum(dim=-1)

        row_indices = torch.arange(count).unsqueeze(0).expand(batch, -1)
        output = weight[row_indices, ind]

        ctx.save_for_backward(input, weight, ind)
        return output

    @staticmethod
    def backward(ctx, grad_output):
        input, weight, ind = ctx.saved_tensors
        grad_input = grad_weight = None

        batch = input.shape[0]
        count = input.shape[1]
        bits = input.shape[2]

        if ctx.needs_input_grad[1]:
            grad_weight = torch.zeros_like(weight)
            ind_p = ind.permute(1, 0)
            grad_output_p = grad_output.permute(1, 0)
            grad_weight.scatter_add_(1, ind_p, grad_output_p)

        if ctx.needs_input_grad[0]:
            row_indices = torch.arange(count).unsqueeze(0).expand(batch, -1)
            grad_input = grad_output * weight[row_indices, ind]
            grad_input = grad_input.unsqueeze(-1).repeat(1, 1, bits)

        return grad_input, grad_weight


class SimpleCNN(nn.Module):
    def __init__(self):
        super(SimpleCNN, self).__init__()
        self.conv1 = nn.Conv2d(1, 10, kernel_size=5)
        self.conv2 = nn.Conv2d(10, 20, kernel_size=5)
        self.bn = nn.BatchNorm1d(320 * 4)
        self.fc1 = nn.Linear(160, 50)
        self.fc2 = nn.Linear(50, 10)
        self.pool = nn.MaxPool2d(2)
        self.relu = nn.ReLU()
        self.weight = nn.Parameter(torch.randn(160, pow(2, 8)))

    def forward(self, x):
        x = self.relu(self.pool(self.conv1(x)))
        x = self.relu((self.conv2(x)))
        x = x.view(-1, 320 * 4)
        x = self.bn(x)
        x = x.view(-1, 160, 8)

        x = Lut.apply(x, self.weight)

        x = self.relu(self.fc1(x))
        x = self.fc2(x)
        return x


class LutGroup(nn.Module):
    def __init__(self, bits, subbits):
        super(LutGroup, self).__init__()
        assert (bits % subbits) == 0
        self.weight = nn.Parameter(torch.randn(bits, pow(2, subbits)))
        self.bits = bits
        self.subbits = subbits

    def forward(self, x):
        batch = x.shape[0]
        x = x.view(batch, -1, self.subbits)
        x = Lut.apply(x, self.weight)
        return x


class LutParallel(nn.Module):
    def __init__(self, bits, number):
        super(LutParallel, self).__init__()
        self.number = number
        self.weight = nn.Parameter(torch.randn(number, pow(2, bits)))

    def forward(self, x):
        x = x.unsqueeze(1).repeat(1, self.number, 1)
        x = Lut.apply(x, self.weight)
        return x


class SimpleBNN(nn.Module):
    def __init__(self):
        super(SimpleBNN, self).__init__()
        # self.w = nn.Parameter(torch.randn(3, 784 * 8))
        # self.b = nn.Parameter(torch.zeros(3, 784 * 8))

        self.w = nn.Parameter(torch.randn(3, 784))
        self.b = nn.Parameter(torch.zeros(3, 784))

        self.lut1 = LutGroup(784 * 8, 8)
        self.lut2 = LutGroup(784, 4)
        self.lut3 = LutGroup(196, 4)
        self.lut4 = LutGroup(49, 7)
        self.lut5 = LutParallel(7, 10)

        self.conv1 = nn.Conv2d(1, 10, kernel_size=5)
        self.conv2 = nn.Conv2d(10, 20, kernel_size=5)
        self.fc1 = nn.Linear(320, 50)
        self.fc2 = nn.Linear(50, 10)
        self.pool = nn.MaxPool2d(2)
        self.relu = nn.ReLU()

    def forward(self, x):
        batch = x.shape[0]
        x = x.view(batch, -1)

        # 变换x [-0.5:0.5] 到 0-255，然后按照二进制展开成8个值
        # x = (x * 256 + 128).clamp(0, 255).to(torch.uint8)
        # xx = torch.arange(7, -1, -1).to(x.device)
        # bits = (x.unsqueeze(-1) >> xx) & 1
        # x = bits.view(batch, -1)
        # x = x.float() - 0.5

        x = (x > 0).float()

        q = x * self.w[0] + self.b[0]
        k = x * self.w[1] + self.b[1]
        v = x * self.w[2] + self.b[2]
        q = q.view(batch, -1, 1)
        k = k.view(batch, 1, -1)
        v = v.view(batch, -1, 1)
        kq = q @ k
        kqv = kq @ v
        kqv = kqv.view(batch, -1, 8)
        x = kqv

        #########################

        # x = (x > 0) << xx
        # x = x.sum(2)
        # x = x.view(batch, 1, 28, 28)
        # x = (x - 128.0) / 256.0

        # x = x.view(batch, 1, 28, 28)
        # x = (x > 0).float()

        # x = self.relu(self.pool(self.conv1(x)))
        # x = self.relu(self.pool((self.conv2(x))))
        # x = x.view(-1, 320)
        # x = self.relu(self.fc1(x))
        # x = self.fc2(x)

        #########################

        x = (x > 0).float()
        x = x.view(batch, 196, 4)

        # x = self.lut1(x)
        x = self.lut2(x)

        x = x.view(-1, 28, 7)
        x = x.permute(0, 2, 1)
        x = x.reshape(-1, 28 * 7)

        x = self.lut3(x)
        x = self.lut4(x)
        x = self.lut5(x)
        return x


torch.autograd.set_detect_anomaly(True)
# model = SimpleCNN().to(device)
model = SimpleBNN().to(device)
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.SGD(model.parameters(), lr=0.001)


def train(epoch):
    model.train()
    for batch_idx, (data, target) in enumerate(train_loader):
        data, target = data.to(device), target.to(device)
        optimizer.zero_grad()
        output = model(data)
        # output = output * 1.0
        # output = F.softmax(output, dim=1)

        # print(output.requires_grad)
        # print(output.grad_fn)

        loss = criterion(output, target)
        loss.backward()
        optimizer.step()

        if batch_idx % 100 == 0:
            print(
                f"Train Epoch: {epoch} [{batch_idx * len(data)}/{len(train_loader.dataset)} "
                f"({100. * batch_idx / len(train_loader):.0f}%)]\tLoss: {loss.item():.6f}"
            )


def test():
    model.eval()
    test_loss = 0
    correct = 0
    with torch.no_grad():
        for data, target in test_loader:
            data, target = data.to(device), target.to(device)
            output = model(data)
            test_loss += criterion(output, target).item()
            pred = output.argmax(dim=1, keepdim=True)
            correct += pred.eq(target.view_as(pred)).sum().item()

    test_loss /= len(test_loader.dataset)
    accuracy = 100.0 * correct / len(test_loader.dataset)
    print(
        f"\nTest set: Average loss: {test_loss:.4f}, Accuracy: {correct}/{len(test_loader.dataset)} "
        f"({accuracy:.0f}%)\n"
    )


for epoch in range(1, 30):
    train(epoch)
    test()

# torch.save(model.state_dict(), "mnist_cnn.pth")
print("Model saved to mnist_cnn.pth")