Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Adding torch.compile compatiblity #9

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
+14 −4
Open

Adding torch.compile compatiblity #9

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 6 additions & 2 deletions scripts/benchmark.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,7 @@
import torch
import numpy as np
import pandas as pd
import e3nn
from e3nn.o3._spherical_harmonics import _spherical_harmonics

from equitriton.sph_harm.bindings import *
Expand Down Expand Up @@ -82,11 +83,14 @@ def e3nn_benchmark(tensor_shape: list[int], device: str | torch.device, l_max: i
joint_tensor[..., 1].contiguous(),
joint_tensor[..., 2].contiguous(),
)
output = _spherical_harmonics(l_max, x, y, z)
e3nn.set_optimization_defaults(jit_script_fx=False)
output = torch.compile(_spherical_harmonics, fullgraph=True, mode="max-autotune")(l_max, x, y, z)
output.backward(gradient=torch.ones_like(output))
# delete references to ensure memory gets cleared
del output
del joint_tensor
e3nn.set_optimization_defaults(jit_script_fx=True) # Turn it back on to avoid any issues



@benchmark(num_steps=args.num_steps, warmup_fraction=args.warmup_fraction)
Expand Down Expand Up @@ -131,4 +135,4 @@ def triton_benchmark(tensor_shape: list[int], device: str | torch.device, l_max:
all_data.append(joint_results)

df = pd.DataFrame(all_data)
df.to_csv(f"{args.device}_lmax{args.l_max}_results.csv", index=False)
df.to_csv(f"{args.device}_lmax{args.l_max}_results.csv", index=False)
5 changes: 4 additions & 1 deletion scripts/measure_numerical_error.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@

import torch
import numpy as np
import e3nn
from e3nn.o3._spherical_harmonics import _spherical_harmonics

from equitriton.sph_harm.bindings import *
Expand Down Expand Up @@ -73,7 +74,8 @@ def compare_e3nn_triton(
joint_tensor[..., 1].contiguous(),
joint_tensor[..., 2].contiguous(),
)
e3nn_output = _spherical_harmonics(l_max, x, y, z)
e3nn.set_optimization_defaults(jit_script_fx=False)
e3nn_output = torch.compile(_spherical_harmonics, fullgraph=True, mode="max-autotune")(l_max, x, y, z)
e3nn_output.backward(gradient=torch.ones_like(e3nn_output))
e3nn_grad = joint_tensor.grad.detach().clone()
joint_tensor.grad = None
Expand All @@ -95,6 +97,7 @@ def compare_e3nn_triton(
# delete intermediate tensors to make sure we don't leak
del e3nn_output
del triton_output
e3nn.set_optimization_defaults(jit_script_fx=True) # Turn it back on to avoid any issues
return (signed_fwd_error, signed_bwd_error)


Expand Down
5 changes: 4 additions & 1 deletion scripts/profile_script.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,7 @@

import torch
from torch.profiler import record_function
import e3nn
from e3nn.o3._spherical_harmonics import _spherical_harmonics

from equitriton.sph_harm.bindings import *
Expand Down Expand Up @@ -74,13 +75,15 @@ def e3nn_benchmark(tensor_shape: list[int], device: str | torch.device, l_max: i
joint_tensor[..., 1].contiguous(),
joint_tensor[..., 2].contiguous(),
)
e3nn.set_optimization_defaults(jit_script_fx=False)
with record_function("forward"):
output = _spherical_harmonics(l_max, x, y, z)
output = torch.compile(_spherical_harmonics, fullgraph=True, mode="max-autotune")(l_max, x, y, z)
with record_function("backward"):
output.backward(gradient=torch.ones_like(output))
# delete references to ensure memory gets cleared
del output
del joint_tensor
e3nn.set_optimization_defaults(jit_script_fx=True) # Turn it back on to avoid any issues


@profile(
Expand Down