finish TsallisDivergenceLoss #114

S1/uucoco_#114/TsallisDivergenceLoss_cuda.py

@@ -0,0 +1,91 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from torch.utils.cpp_extension import load_inline
+
+cuda_source = """
+#include <torch/extension.h>
+#include <cuda_runtime.h>
+
+__global__ void tsallis_divergence_kernel(
+    const float* __restrict__ p,
+    const float* __restrict__ q,
+    float* __restrict__ output,
+    int num_classes,
+    float q_param
+) {
+    extern __shared__ float sdata[];
+
+    int tid = threadIdx.x;
+    int bid = blockIdx.x;
+
+    int row_offset = bid * num_classes;
+    float local_sum = 0.0f;
+
+    for (int i = tid; i < num_classes; i += blockDim.x) {
+        float p_val = p[row_offset + i];
+        float q_val = q[row_offset + i];
+
+        local_sum += powf(p_val, q_param) * powf(q_val, 1.0f - q_param);
+    }
+
+    sdata[tid] = local_sum;
+    __syncthreads();
+
+    for (int s = blockDim.x / 2; s > 0; s >>= 1) {
+        if (tid < s) {
+            sdata[tid] += sdata[tid + s];
+        }
+        __syncthreads();
+    }
+
+    if (tid == 0) {
+        float sum_val = sdata[0];
+        output[bid] = (sum_val - 1.0f) / (q_param - 1.0f);
+    }
+}
+
+torch::Tensor tsallis_divergence_cuda(torch::Tensor p, torch::Tensor q, float q_param) {
+    int batch_size = p.size(0);
+    int num_classes = p.size(1);
+
+    auto output = at::empty({batch_size}, p.options());
+
+    int threads = 256;
+    int blocks = batch_size;
+    int shared_mem = threads * sizeof(float);
+
+    tsallis_divergence_kernel<<<blocks, threads, shared_mem>>>(
+        p.data_ptr<float>(),
+        q.data_ptr<float>(),
+        output.data_ptr<float>(),
+        num_classes,
+        q_param
+    );
+
+    return output.mean();
+}
+"""
+
+cpp_source = """
+torch::Tensor tsallis_divergence_cuda(torch::Tensor p, torch::Tensor q, float q_param);
+"""
+
+tsallis_divergence_loss = load_inline(
+    name="tsallis_divergence_loss",
+    cpp_sources=cpp_source,
+    cuda_sources=cuda_source,
+    functions=["tsallis_divergence_cuda"],
+    verbose=False
+)
+
+
+class ModelNew(nn.Module):
+    def __init__(self, q=0.5):
+        super(ModelNew, self).__init__()
+        self.q = q
+
+    def forward(self, p, target):
+        p_prob = F.softmax(p, dim=1)
+        q_prob = F.softmax(target, dim=1)
+        return tsallis_divergence_loss.tsallis_divergence_cuda(p_prob, q_prob, self.q)

S1/uucoco_#114/TsallisDivergenceLoss_torch.py

@@ -0,0 +1,32 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+
+class Model(nn.Module):
+    def __init__(self, q=0.5):
+        super(Model, self).__init__()
+        self.q = q
+
+    def forward(self, p, target):
+        p_prob = F.softmax(p, dim=1)
+        q_prob = F.softmax(target, dim=1)
+
+        sum_term = torch.sum((p_prob ** self.q) * (q_prob ** (1.0 - self.q)), dim=1)
+        loss = (sum_term - 1.0) / (self.q - 1.0)
+
+        return loss.mean()
+
+
+batch_size = 32
+num_classes = 1000
+
+
+def get_inputs():
+    p = torch.randn(batch_size, num_classes, requires_grad=True)
+    target = torch.randn(batch_size, num_classes)
+    return [p, target]
+
+
+def get_init_inputs():
+    return [0.5]

S1/uucoco_#114/prompt.txt

@@ -0,0 +1,131 @@
+You write custom CUDA kernels to replace the pytorch operators in the given GeGLU architecture to get speedups.
+
+You have complete freedom to choose the set of operators you want to replace. You may make the decision to replace some operators with custom CUDA kernels and leave others unchanged. You may replace multiple operators with custom implementations, consider operator fusion opportunities (combining multiple operators into a single kernel, for example, combining chunk+gelu+elementwise_mul), or algorithmic changes (such as optimized memory access patterns). You are only limited by your imagination.
+Technologies Used in This Code
+Core Libraries & Frameworks
+PyTorch: Deep learning framework
+
+CUDA: NVIDIA's parallel computing platform for GPU acceleration
+
+C++: For high-performance kernel implementation
+
+PyTorch Specific Components
+torch.nn.Module: Base class for neural network modules
+
+torch.nn.functional.F.softmax: Softmax activation function
+
+torch.utils.cpp_extension.load_inline: For inline compilation of CUDA/C++ extensions
+
+PyTorch Tensors: Multi-dimensional arrays with automatic differentiation
+
+CUDA/C++ Implementation Details
+CUDA Kernels: Custom GPU kernel (tsallis_divergence_kernel)
+
+CUDA Math Functions: powf() for floating-point exponentiation
+
+Parallel Reduction: Tree-based reduction using shared memory
+
+Shared Memory: Using __shared__ for inter-thread communication
+
+Block-Level Parallelism: One CUDA block per batch element
+
+Grid-Stride Loops: Efficient memory access within each row
+
+Mathematical Components
+Tsallis Divergence: Non-extensive entropy-based divergence measure
+
+Power Operations: powf(p, q) * powf(q_dist, 1-q) formulation
+
+Linear Normalization: (sum - 1) / (q - 1) scaling
+
+Statistical Distance: Measures difference between probability distributions
+
+q-Parameter: Controls divergence properties (q ≠ 1)
+
+Memory & Parallelism Patterns
+Per-Batch Block Assignment: One CUDA block processes one batch element
+
+Shared Memory Reduction: Tree reduction within thread blocks
+
+Row-Wise Processing: Threads parallelize across class dimensions within rows
+
+Batch Independence: Parallel processing across batch dimension
+
+Optimization Techniques
+Grid-Stride Loops: Threads process multiple elements within their assigned row
+
+Shared Memory Efficiency: Single buffer for intermediate sums
+
+Fused Computation: Complete Tsallis divergence calculation per batch element
+
+Numerical Stability: Linear scaling after summation
+
+Coalesced Memory Access: Sequential memory access patterns
+
+Performance Features
+Massive Parallelization: GPU acceleration for divergence computation
+
+Memory Efficiency: Shared memory reuse for reduction operations
+
+Scalable Design: Efficient for varying batch sizes and class counts
+
+Minimal Synchronization: Single __syncthreads() call per reduction
+
+Batch Mean Computation: Final averaging performed on CPU
+
+Unique Implementation Aspects
+q-Parameter Naming: Note: Uses q_param (not to confuse with input q tensor)
+
+Linear Scaling: Tsallis divergence uses linear rather than logarithmic scaling
+
+Power Product: Similar to Rényi but with different normalization
+
+Per-Sample Output: Each batch element gets its own divergence value
+
+Non-Extensive Statistics: Based on Tsallis entropy formulation
+
+Comparison with Similar Divergences
+vs Rényi: Uses linear (sum-1)/(q-1) instead of logarithmic log(sum)/(q-1)
+
+vs Alpha Divergence: Similar power structure but different normalization
+
+Parameter Range: Typically q > 0, q ≠ 1 for proper divergence definition
+
+
+
+
+
+
+Here's an example to show you the syntax of inline embedding custom CUDA operators in torch: The example given architecture is:
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+
+class Model(nn.Module):
+    def __init__(self, q=0.5):
+        super(Model, self).__init__()
+        self.q = q
+
+    def forward(self, p, target):
+        p_prob = F.softmax(p, dim=1)
+        q_prob = F.softmax(target, dim=1)
+
+        sum_term = torch.sum((p_prob ** self.q) * (q_prob ** (1.0 - self.q)), dim=1)
+        loss = (sum_term - 1.0) / (self.q - 1.0)
+
+        return loss.mean()
+
+
+batch_size = 32
+num_classes = 1000
+
+
+def get_inputs():
+    p = torch.randn(batch_size, num_classes, requires_grad=True)
+    target = torch.randn(batch_size, num_classes)
+    return [p, target]
+
+
+def get_init_inputs():
+    return [0.5]

S1/uucoco_#114/run_code.py

@@ -0,0 +1,77 @@
+###########################################################
+# 性能和精度验证程序
+###########################################################
+import torch
+import torch.nn as nn
+import time
+from TsallisDivergenceLoss_torch import Model, get_inputs, get_init_inputs
+from TsallisDivergenceLoss_cuda  import ModelNew
+
+
+def run_benchmark():
+    # 检查 CUDA 是否可用
+    if not torch.cuda.is_available():
+        print("CUDA 不可用，请确保您有可用的 NVIDIA GPU 并已正确安装 PyTorch CUDA 版本。")
+        return
+    else:
+        device = torch.device("cuda")
+
+    # 初始化模型
+    init_inputs = get_init_inputs()
+    init_inputs = [
+        x.cuda(device=device) if isinstance(x, torch.Tensor) else x for x in init_inputs
+    ]
+    inputs = get_inputs()
+    inputs = [
+        x.cuda(device=device) if isinstance(x, torch.Tensor) else x for x in inputs
+    ]
+
+    torch_model = Model(*init_inputs).cuda()
+    cuda_model = ModelNew(*init_inputs).cuda()
+
+    torch_model.eval()
+    cuda_model.eval()
+
+    print("-------------------- 精度对齐验证 --------------------")
+    with torch.no_grad():
+        output_torch = torch_model(*inputs)
+        output_cuda = cuda_model(*inputs)
+
+    precision_flag = torch.allclose(output_torch, output_cuda, rtol=1e-03)
+    if precision_flag:
+        print("✅ 精度对齐：两个模型的输出结果非常接近。")
+    else:
+        print("❌ 精度不一致！")
+
+    print("\n-------------------- 性能加速比测试 --------------------")
+    num_iterations = 100
+
+    # PyTorch 模型计时
+    torch.cuda.synchronize()
+    start_time = time.time()
+    for _ in range(num_iterations):
+        _ = torch_model(*inputs)
+    torch.cuda.synchronize()
+    torch_time = (time.time() - start_time) / num_iterations
+
+    # 自定义 CUDA 内核计时
+    torch.cuda.synchronize()
+    start_time = time.time()
+    for _ in range(num_iterations):
+        _ = cuda_model(*inputs)
+    torch.cuda.synchronize()
+    cuda_time = (time.time() - start_time) / num_iterations
+
+    print(f"PyTorch torch.relu 平均执行时间: {torch_time:.6f} 秒")
+    print(f"自定义 CUDA 内核 平均执行时间: {cuda_time:.6f} 秒")
+    speedup = 0
+    if cuda_time > 0:
+        speedup = torch_time / cuda_time
+        print(f"加速比 (Speedup): {speedup:.2f}x")
+    else:
+        print("CUDA 内核执行时间为0，无法计算加速比。")
+    return precision_flag, speedup
+
+
+if __name__ == "__main__":
+    precision_flag, speedup = run_benchmark()