Update torch-ext/triton_kernels/matmul_ogs_details/opt_flags.py (#3)

- Update torch-ext/triton_kernels/matmul_ogs_details/opt_flags.py (87f694d54e44f47a7aab2c88d004610114322262)

torch-ext/triton_kernels/matmul_ogs_details/opt_flags.py

@@ -211,7 +211,8 @@ def make_default_opt_flags_nvidia(
         ns = opt_flags_nvidia.compute_num_stages(*compute_num_stages_args, ep, epilogue_effective_itemsize)
         if ns > num_stages:
             epilogue_subtile, num_stages = ep, ns
-    assert num_stages >= 1
+    # removed due to https://huggingface.co/kernels-community/triton_kernels/discussions/1
+    # assert num_stages >= 1
     if constraints.get("num_stages", None):
         num_stages = constraints["num_stages"]