Added MoE file for the Hunyuan 3D 2.1 Model

comfy/ldm/hunyuan3d/model/moe.py

@@ -0,0 +1,203 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+class GELU(nn.Module):
+
+    def __init__(self, dim_in: int, dim_out: int):
+        super().__init__()
+        self.proj = nn.Linear(dim_in, dim_out)
+
+    def gelu(self, gate: torch.Tensor) -> torch.Tensor:
+
+        if gate.device.type == "mps":
+            return F.gelu(gate.to(dtype = torch.float32)).to(dtype = gate.dtype)
+        
+        return F.gelu(gate)
+
+    def forward(self, hidden_states):
+
+        hidden_states = self.proj(hidden_states)
+        hidden_states = self.gelu(hidden_states)
+
+        return hidden_states
+
+class FeedForward(nn.Module):
+
+    def __init__(self, dim: int, dim_out = None, mult: int = 4,
+                dropout: float = 0.0, inner_dim = None):
+        
+        super().__init__()
+        if inner_dim is None:
+            inner_dim = int(dim * mult)
+
+        dim_out = dim_out if dim_out is not None else dim
+
+        act_fn = GELU(dim, inner_dim)
+
+        self.net = nn.ModuleList([])
+        self.net.append(act_fn)
+
+        self.net.append(nn.Dropout(dropout))
+        self.net.append(nn.Linear(inner_dim, dim_out))
+
+    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+        for module in self.net:
+            hidden_states = module(hidden_states)
+        return hidden_states
+
+class AddAuxLoss(torch.autograd.Function):
+
+    @staticmethod
+    def forward(ctx, x, loss):
+        # do nothing in forward (no computation)
+        assert loss.numel() == 1
+        ctx.requires_aux_loss = loss.requires_grad  
+        ctx.dtype = loss.dtype
+
+        return x
+    
+    @staticmethod
+    def backward(ctx, grad_output):
+        # add the aux loss gradients
+        grad_loss = None
+        # put the aux grad the same as the main grad loss
+        # aux grad contributes equally
+        if ctx.requires_aux_loss:
+            grad_loss = torch.ones(1, dtype = ctx.dtype, device = grad_output.device)
+
+        return grad_output, grad_loss
+    
+class MoEGate(nn.Module):
+
+    def __init__(self, embed_dim, num_experts=16, num_experts_per_tok=2, aux_loss_alpha=0.01):
+
+        super().__init__()
+        self.top_k = num_experts_per_tok
+        self.n_routed_experts = num_experts
+
+        self.alpha = aux_loss_alpha
+        self.seq_aux = False
+
+        self.gating_dim = embed_dim
+        self.weight = nn.Parameter(torch.empty((self.n_routed_experts, self.gating_dim)))
+
+    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+
+        # flatten hidden states
+        hidden_states = hidden_states.view(-1, hidden_states.size(-1))
+
+        # get logits and pass it to softmax
+        logits = F.linear(hidden_states, self.weight, bias = None)
+        scores = logits.softmax(dim = -1)
+
+        topk_weight, topk_idx = torch.topk(scores, k = self.top_k, dim = -1, sorted = False)
+
+        if self.training and self.alpha > 0.0:
+            scores_for_aux = scores
+
+            # used bincount instead of one hot encoding
+            counts = torch.bincount(topk_idx.view(-1), minlength = self.n_routed_experts).float()
+            ce = counts / topk_idx.numel()  # normalized expert usage
+
+            # mean expert score
+            Pi = scores_for_aux.mean(0)
+
+            # expert balance loss
+            aux_loss = (Pi * ce * self.n_routed_experts).sum() * self.alpha
+        else:
+            aux_loss = None
+
+        return topk_idx, topk_weight, aux_loss
+
+class MoEBlock(nn.Module):
+    def __init__(self, dim, num_experts: int = 6, moe_top_k: int = 2, dropout: float = 0.0, ff_inner_dim: int = None):
+        super().__init__()
+
+        self.moe_top_k = moe_top_k
+        self.num_experts = num_experts
+
+        self.experts = nn.ModuleList([
+            FeedForward(dim, dropout = dropout, inner_dim = ff_inner_dim)
+            for _ in range(num_experts)
+        ])
+
+        self.gate = MoEGate(dim, num_experts = num_experts, num_experts_per_tok = moe_top_k)
+        self.shared_experts = FeedForward(dim, dropout = dropout, inner_dim = ff_inner_dim)
+
+    def forward(self, hidden_states) -> torch.Tensor:
+
+        identity = hidden_states
+        orig_shape = hidden_states.shape
+        topk_idx, topk_weight, aux_loss = self.gate(hidden_states)
+
+        hidden_states = hidden_states.view(-1, hidden_states.shape[-1])
+        flat_topk_idx = topk_idx.view(-1)
+        
+        if self.training:
+
+            hidden_states = hidden_states.repeat_interleave(self.moe_top_k, dim = 0)
+            y = torch.empty_like(hidden_states, dtype = hidden_states.dtype)
+
+            for i, expert in enumerate(self.experts): 
+                tmp = expert(hidden_states[flat_topk_idx == i])
+                y[flat_topk_idx == i] = tmp.to(hidden_states.dtype)
+
+            y = (y.view(*topk_weight.shape, -1) * topk_weight.unsqueeze(-1)).sum(dim = 1)
+            y =  y.view(*orig_shape)
+
+            y = AddAuxLoss.apply(y, aux_loss)
+        else:
+            y = self.moe_infer(hidden_states, flat_expert_indices = flat_topk_idx,flat_expert_weights = topk_weight.view(-1, 1)).view(*orig_shape)
+
+        y = y + self.shared_experts(identity)
+
+    @torch.no_grad()
+    def moe_infer(self, x, flat_expert_indices, flat_expert_weights):
+        
+        expert_cache = torch.zeros_like(x) 
+        idxs = flat_expert_indices.argsort()
+
+        # no need for .numpy().cpu() here
+        tokens_per_expert = flat_expert_indices.bincount().cumsum(0)
+        token_idxs = idxs // self.moe_top_k 
+        
+        for i, end_idx in enumerate(tokens_per_expert):
+
+            start_idx = 0 if i == 0 else tokens_per_expert[i-1]
+
+            if start_idx == end_idx:
+                continue
+
+            expert = self.experts[i]
+            exp_token_idx = token_idxs[start_idx:end_idx]
+
+            expert_tokens = x[exp_token_idx]
+            expert_out = expert(expert_tokens)
+
+            expert_out.mul_(flat_expert_weights[idxs[start_idx:end_idx]]) 
+
+            # use index_add_ with a 1-D index tensor directly avoids building a large [N, D] index map and extra memcopy required by scatter_reduce_
+            # + avoid dtype conversion
+            expert_cache.index_add_(0, exp_token_idx, expert_out)
+
+        return expert_cache
+
+def test_moe():
+
+    torch.manual_seed(2025)
+    import time
+
+    start = time.time()
+
+    moe_gate = MoEGate(512)
+    moe_gate(torch.rand(1, 71, 512))
+
+    moe_block = MoEBlock(512)
+    moe_block(torch.rand(1, 77, 512))
+
+    timing = time.time() - start
+    print(timing)
+
+if __name__ == "__main__":
+    test_moe()