bev-project/MEMORY_ANALYSIS.py

#!/usr/bin/env python
"""
分析BEVFusion Phase 4B显存占用情况
"""

import torch
import numpy as np

def calculate_memory_usage():
    print("=== BEVFusion Phase 4B 显存占用分析 ===\n")

    # ========== 基础参数 (优化后) ==========
    batch_size = 2  # samples_per_gpu (优化: 从4减少到2)
    bev_height = 360  # BEV高度 (优化: 从598减少到360)
    bev_width = 360   # BEV宽度 (优化: 从598减少到360)
    bev_channels = 512  # BEV特征通道数

    # ========== 1. BEV特征显存占用 ==========
    print("1. BEV特征显存占用:")
    bev_size = batch_size * bev_channels * bev_height * bev_width
    bev_memory_mb = bev_size * 4 / (1024 * 1024)  # float32 = 4 bytes
    print(f"BEV特征: {bev_memory_mb:.1f} MB")

    # ========== 2. 多尺度特征显存占用 ==========
    print("\n2. 多尺度特征显存占用:")
    scales = [180, 360, 540]  # 三尺度 (优化: 从[299, 598, 996]改为[180, 360, 540])
    total_multi_scale = 0
    for scale in scales:
        scale_size = batch_size * bev_channels * scale * scale
        scale_memory_mb = scale_size * 4 / (1024 * 1024)
        total_multi_scale += scale_memory_mb
        print(f"  {scale}x{scale}: {scale_memory_mb:.1f} MB")
    print(f"多尺度总计: {total_multi_scale:.1f} MB")

    # ========== 3. Transformer解码器显存占用 ==========
    print("\n3. Transformer解码器显存占用:")
    # 假设6个类别，每个类别独立处理
    num_classes = 6
    transformer_features = batch_size * num_classes * bev_channels * bev_height * bev_width
    transformer_memory_mb = transformer_features * 4 / (1024 * 1024)
    print(f"Transformer解码器: {transformer_memory_mb:.1f} MB")

    # ========== 4. 3D检测分支显存占用 ==========
    print("\n4. 3D检测分支显存占用:")
    # TransFusion头输出特征
    det_channels = 256
    det_queries = 900  # 查询数量
    det_features = batch_size * det_queries * det_channels
    det_memory_mb = det_features * 4 / (1024 * 1024)
    print(f"3D检测特征: {det_memory_mb:.1f} MB")

    # ========== 5. 中间特征和梯度显存占用 ==========
    print("\n5. 中间特征和梯度占用:")

    # Swin-Transformer特征
    swin_stages = [
        (batch_size, 128, 128, 352),  # stage1
        (batch_size, 256, 64, 176),   # stage2
        (batch_size, 512, 32, 88),    # stage3
        (batch_size, 1024, 16, 44),   # stage4
    ]

    swin_memory = 0
    for stage_name, shape in zip(['stage1', 'stage2', 'stage3', 'stage4'], swin_stages):
        stage_size = np.prod(shape)
        stage_memory_mb = stage_size * 4 / (1024 * 1024)
        swin_memory += stage_memory_mb
        print(f"  {stage_name}: {stage_memory_mb:.1f} MB")
    print(f"Swin-Transformer总计: {swin_memory:.1f} MB")

    # ========== 6. 梯度占用 (训练时) ==========
    print("\n6. 梯度占用 (训练时):")
    # 假设所有参数都需要梯度，梯度占用约等于参数占用
    total_params_memory = (
        bev_memory_mb + total_multi_scale + transformer_memory_mb +
        det_memory_mb + swin_memory
    )
    gradient_memory_mb = total_params_memory  # 粗略估计
    print(f"梯度占用: {gradient_memory_mb:.1f} MB")

    # ========== 7. 总计显存占用 ==========
    print("\n7. 总计显存占用:")

    # 前向传播
    forward_memory = (
        bev_memory_mb + total_multi_scale + transformer_memory_mb +
        det_memory_mb + swin_memory
    )

    # 训练时的额外占用 (梯度 + 优化器状态)
    training_overhead = forward_memory * 2  # 梯度 + Adam状态

    total_training_memory = forward_memory + training_overhead

    print(f"前向传播: {forward_memory:.1f} MB")
    print(f"训练总计: {total_training_memory:.1f} MB")

    print("\n8. 优化建议:")

    # 计算每个部分的占比
    components = {
        'BEV特征': bev_memory_mb,
        '多尺度特征': total_multi_scale,
        'Transformer解码器': transformer_memory_mb,
        '3D检测': det_memory_mb,
        'Swin-Transformer': swin_memory
    }

    total_components = sum(components.values())
    print("各组件显存占比:")
    for name, memory in components.items():
        percentage = (memory / total_components) * 100
        print(f"  {name}: {percentage:.1f}%")

    print("\n显存优化建议:")
    print("1. 降低batch_size: 当前4 → 建议2或1")
    print("2. 减少BEV分辨率: 598×598 → 360×360 (减少67%)")
    print("3. 减少多尺度数量: 3尺度 → 2尺度")
    print("4. 使用梯度累积: 模拟更大的batch_size")
    print("5. 使用FP16训练: 减少50%显存占用")
    print("6. 减少Transformer层数或通道数")

if __name__ == '__main__':
    calculate_memory_usage()