+13
−4
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/153233
Note: Links to docs will display an error until the docs builds have been completed. This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This comment was marked as outdated.
This comment was marked as outdated.
|
@pytorchbot label "module: error checking" |
|
@pytorchbot label "module: norms and normalization" |
Skylion007
reviewed
May 9, 2025
e402021 to
136ab31
Compare
import torch
print(f"PyTorch Version: {torch.__version__}")
# Common parameters for torch.batch_norm
weight_param = None
bias_param = None
is_training_param = True # Error occurs with True or False
momentum_param = 0.1
eps_param = 1e-5
cudnn_enabled_param = True # Also occurs with False on GPU
# --- Scenario 1: running_mean is Tensor, running_var is None ---
print("\n--- Scenario 1: running_mean is Tensor, running_var is None ---")
# Input tensor
input_tensor_shape = (3, 4, 5) # N, C, D*
num_features = input_tensor_shape[1]
# CPU
print(" CPU (Scenario 1):")
try:
input_tensor_cpu = torch.randn(input_tensor_shape)
running_mean_param_cpu = torch.randn(num_features)
running_var_param_cpu = None
torch.batch_norm(
input_tensor_cpu,
weight_param,
bias_param,
running_mean_param_cpu,
running_var_param_cpu,
is_training_param,
momentum_param,
eps_param,
cudnn_enabled_param
)
print(" CPU: Error not triggered.")
except ValueError as e:
print(f" CPU Error: {e}")
if "Expected has_running_mean == has_running_var to be true, but got false" in str(e):
print(" CPU: Successfully triggered the target error (unexpected based on current behavior).")
# GPU
if torch.cuda.is_available():
print(" GPU (Scenario 1):")
try:
input_tensor_gpu = torch.randn(input_tensor_shape).cuda()
running_mean_param_gpu = torch.randn(num_features).cuda()
running_var_param_gpu = None
torch.batch_norm(
input_tensor_gpu,
weight_param,
bias_param,
running_mean_param_gpu,
running_var_param_gpu,
is_training_param,
momentum_param,
eps_param,
cudnn_enabled_param
)
print(" GPU: Error not triggered (unexpected for this specific error message).")
except ValueError as e:
print(f" GPU Error: {e}")
if "Expected has_running_mean == has_running_var to be true, but got false" in str(e):
print(" GPU: Successfully triggered the target error.")
else:
print(" GPU (Scenario 1): CUDA not available, skipping GPU test.")
# --- Scenario 2: running_mean is None, running_var is Tensor ---
print("\n--- Scenario 2: running_mean is None, running_var is Tensor ---")
# CPU
print(" CPU (Scenario 2):")
try:
input_tensor_cpu = torch.randn(input_tensor_shape)
running_mean_param_cpu = None
running_var_param_cpu = torch.randn(num_features)
torch.batch_norm(
input_tensor_cpu,
weight_param,
bias_param,
running_mean_param_cpu,
running_var_param_cpu,
is_training_param,
momentum_param,
eps_param,
cudnn_enabled_param
)
print(" CPU: Error not triggered.")
except ValueError as e:
print(f" CPU Error: {e}")
if "Expected has_running_mean == has_running_var to be true, but got false" in str(e):
print(" CPU: Successfully triggered the target error (unexpected based on current behavior).")
# GPU
if torch.cuda.is_available():
print(" GPU (Scenario 2):")
try:
input_tensor_gpu = torch.randn(input_tensor_shape).cuda()
running_mean_param_gpu = None
running_var_param_gpu = torch.randn(num_features).cuda()
torch.batch_norm(
input_tensor_gpu,
weight_param,
bias_param,
running_mean_param_gpu,
running_var_param_gpu,
is_training_param,
momentum_param,
eps_param,
cudnn_enabled_param
)
print(" GPU: Error not triggered (unexpected for this specific error message).")
except ValueError as e:
print(f" GPU Error: {e}")
if "Expected has_running_mean == has_running_var to be true, but got false" in str(e):
print(" GPU: Successfully triggered the target error.")
else:
print(" GPU (Scenario 2): CUDA not available, skipping GPU test.")Change to use ValueError. --- Scenario 2: running_mean is None, running_var is Tensor --- |
|
Could you help review this pr? @Skylion007 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #153137
Aligned batch_norm_cpu_out assertion to batch_norm_cuda_out.
cc @malfet