Update modeling_rwkv6qwen2.py

added check for fla import requirement

modeling_rwkv6qwen2.py

@@ -204,8 +204,14 @@ class RWKV6State(Cache):
     #         self.key_cache[layer_idx] = self.key_cache[layer_idx][indices, ...]
     #         self.value_cache[layer_idx] = self.value_cache[layer_idx][indices, ...]
 
-from fla.ops.gla.chunk import chunk_gla
-from fla.ops.gla.fused_recurrent import fused_recurrent_gla
+try:
+    #from fla.ops.gla.chunk import chunk_gla
+    from fla.ops.gla.fused_recurrent import fused_recurrent_gla
+except ImportError:
+    print("Required module is not installed. Please install it using the following commands:")
+    print("pip install -U git+https://github.com/sustcsonglin/flash-linear-attention")
+    print("Additionally, ensure you have at least version 2.2.0 of Triton installed:")
+    print("pip install triton>=2.2.0")
 
 class RWKV6Attention(nn.Module):
     def __init__(self, config, layer_idx: Optional[int] = None):