[Nested Tensor / subclasses] view(-1) and splitting dimensions with view() failures about pytorch HOT 2 OPEN

Issue 2 makes me sad. AFAICT the symbol is not being simplified out because it's part of guard expressions:

• Eq(s5*s6*s7, 60) from the view decomp. Solving for s7 gives Eq(s7, 60/(s5*s6)), and then the logic attempts to replace s7 with 60/(s5*s6), but the latter doesn't have is_integer=True, so this branch is skipped:
  

  pytorch/torch/fx/experimental/symbolic_shapes.py

  Lines 4947 to 4951 in 2e5366f

  	if r is not None and all(t.is_integer for t in sympy.preorder_traversal(r[1])):
  	new_var = self._find(r[1])
  	ok = len(free_unbacked_symbols(new_var)) == 0
  	if ok:
  	self._set_replacement(cast(sympy.Symbol, free[0]), new_var, "solve")

• Eq(s7, (15//s6)) doesn't allow for symbol replacement either; it throws here:
  

  pytorch/torch/fx/experimental/symbolic_shapes.py

  Lines 4913 to 4914 in 2e5366f

  	if len(floor_div_atoms) > 0 and any(a.divisor != 1 for a in floor_div_atoms):
  	raise NotImplementedError

TL;DR: can't do symbol replacement for any of the guards so the ephemeral symbol created during view replay can't be simplified out

from pytorch.

Here we know that s5, s6 and s7 divide 60 evenly. Thus, the replacement should be allowed. I would probably try a fix along those lines.

However, I've generally found it is better to express things in terms of s5s6s7, instead of trying to replace s7 with 60/(s5*s6), so it would be good to understand why the replacement is preferred as well.

from pytorch.