torch.export failed with dynamic batch for simple convolution model #118289
Description
🐛 Describe the bug
import torch
from torch.export import export
# Simple module for demonstration
class M(torch.nn.Module):
def __init__(self) -> None:
super().__init__()
self.conv = torch.nn.Conv2d(
in_channels=3, out_channels=32
, kernel_size=3, padding=1
)
self.relu = torch.nn.ReLU()
self.maxpool = torch.nn.MaxPool2d(kernel_size=3)
def forward(self, x: torch.Tensor, *, constant=None) -> torch.Tensor:
a = self.conv(x)
a.add_(constant)
return self.maxpool(self.relu(a))
example_args = (torch.randn(2, 3, 256, 256),)
const = torch.ones(2, 32, 256, 256)
example_kwargs = {"constant": const}
constraints = [dynamic_dim(example_args[0], 0), dynamic_dim(const, 0),
dynamic_dim(const, 0) == dynamic_dim(example_args[0], 0),
]
exported_program: torch.export.ExportedProgram = export(
M(), args=example_args, kwargs=example_kwargs, constraints=constraints
)
print(exported_program)**Error message:**
During handling of the above exception, another exception occurred:
UserError Traceback (most recent call last)
[/usr/local/lib/python3.10/dist-packages/torch/_export/__init__.py](https://localhost:8080/#) in export(f, args, kwargs, constraints, preserve_module_call_signature)
278 )
279 except (ConstraintViolationError, ValueRangeError) as e:
--> 280 raise UserError(UserErrorType.CONSTRAIN_VIOLATION, str(e))
281 except GuardOnDataDependentSymNode as e:
282 raise UserError(
UserError: Constraints violated!
1. Not all values of L['x'].size()[0] in the specified range satisfy the generated guard L['x'].size()[0] < 16. For more information about why this guard was generated, run with TORCH_LOGS=dynamic.
2. Not all values of L['x'].size()[0] in the specified range satisfy the generated guard L['x'].size()[0] < 16. For more information about why this guard was generated, run with TORCH_LOGS=dynamic.
The following dimensions have been specialized and CANNOT be dynamic.
def specializations(x: torch.Tensor, *, constant=None):
# x:
assert x.size()[1] == 3
assert x.size()[2] == 256
assert x.size()[3] == 256
# constant:
assert constant.size()[1] == 32
assert constant.size()[2] == 256
assert constant.size()[3] == 256
The following dimensions CAN be dynamic.
Please use the following code to specify the constraints they must satisfy:
def specify_constraints(x: torch.Tensor, *, constant=None):
return [
# x:
dynamic_dim(x, 0) < 16,
# constant:
dynamic_dim(constant, 0) == dynamic_dim(x, 0),
]
Versions
Collecting environment information...
PyTorch version: 2.1.0+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A
OS: Ubuntu 22.04.3 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: 14.0.0-1ubuntu1.1
CMake version: version 3.27.9
Libc version: glibc-2.35
Python version: 3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0] (64-bit runtime)
Python platform: Linux-6.1.58+-x86_64-with-glibc2.35
Is CUDA available: False
CUDA runtime version: 12.2.140
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: Could not collect
Nvidia driver version: Could not collect
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.8.9.6
/usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.9.6
/usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.9.6
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.9.6
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.9.6
/usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.9.6
/usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.9.6
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True
CPU:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 46 bits physical, 48 bits virtual
Byte Order: Little Endian
CPU(s): 2
On-line CPU(s) list: 0,1
Vendor ID: GenuineIntel
Model name: Intel(R) Xeon(R) CPU @ 2.20GHz
CPU family: 6
Model: 79
Thread(s) per core: 2
Core(s) per socket: 1
Socket(s): 1
Stepping: 0
BogoMIPS: 4400.29
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopology nonstop_tsc cpuid tsc_known_freq pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single ssbd ibrs ibpb stibp fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm rdseed adx smap xsaveopt arat md_clear arch_capabilities
Hypervisor vendor: KVM
Virtualization type: full
L1d cache: 32 KiB (1 instance)
L1i cache: 32 KiB (1 instance)
L2 cache: 256 KiB (1 instance)
L3 cache: 55 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0,1
Vulnerability Gather data sampling: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Mitigation; PTE Inversion
Vulnerability Mds: Vulnerable; SMT Host state unknown
Vulnerability Meltdown: Vulnerable
Vulnerability Mmio stale data: Vulnerable
Vulnerability Retbleed: Vulnerable
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Vulnerable
Vulnerability Spectre v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers
Vulnerability Spectre v2: Vulnerable, IBPB: disabled, STIBP: disabled, PBRSB-eIBRS: Not affected
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Vulnerable
Versions of relevant libraries:
[pip3] numpy==1.23.5
[pip3] torch==2.1.0+cu121
[pip3] torchaudio==2.1.0+cu121
[pip3] torchdata==0.7.0
[pip3] torchsummary==1.5.1
[pip3] torchtext==0.16.0
[pip3] torchvision==0.16.0+cu121
[pip3] triton==2.1.0
[conda] Could not collect
cc @avikchaudhuri @gmagogsfm @zhxchen17 @tugsbayasgalan @angelayi @suo @ydwu4