Inconsistent Behavior with transforms.v2 for Multiple Arguments

🐛 Describe the bug

I've been testing various transforms.v2 and noticed an inconsistency:

When passing multiple PIL.Image arguments, the transformation is applied to all of them simultaneously, which is the expected behavior.
However, when passing multiple torch.Tensor arguments, only the first argument is transformed, while the others remain unchanged. This issue isn't just with Resize; it also occurs with other transforms like Normalize. Wouldn't it be more intuitive and consistent if transforms were applied to all provided arguments, whether they are PIL.Image or torch.Tensor? This would lead to a more predictable and natural workflow, especially when dealing with multiple tensors.

Expected Behavior:

The transform should be applied to all provided arguments, regardless of their type (PIL.Image or torch.Tensor), in a consistent manner.

This inconsistency becomes especially confusing when using Compose to chain multiple transforms. The behavior differs depending on whether the input is a PIL.Image or a torch.Tensor, which can lead to unexpected results and confusion.

import torch
import torchvision.transforms.v2 as v2
import PIL.Image as Image

# PIL Image
a = Image.new("L", (2, 2), (1))
b, c = a.copy(), a.copy()

print('-- (pil.image) resize applied --')
print(a.size, b.size, c.size)
a, b, c = v2.Resize(1)(a, b, c)
print(' >> ')
print(a.size, b.size, c.size)

# torch
a = torch.ones([1, 2, 2])
b, c = a.clone(), a.clone()

print('-- (torch) resize applied --')
print(a.shape, b.shape, c.shape)
print(' >> ')
a, b, c = v2.Resize(1)(a, b, c)
print(a.shape, b.shape, c.shape)

-- (pil.image) resize applied --
(2, 2) (2, 2) (2, 2)
 >> 
(1, 1) (1, 1) (1, 1)
-- (torch) resize applied --
torch.Size([1, 2, 2]) torch.Size([1, 2, 2]) torch.Size([1, 2, 2])
 >> 
torch.Size([1, 1, 1]) torch.Size([1, 2, 2]) torch.Size([1, 2, 2])

OS: Ubuntu 20.04.6 LTS PyTorch: 2.3.1 Torchvison: 0.18.1

Versions

PyTorch version: 2.3.1 Is debug build: False CUDA used to build PyTorch: 11.8 ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.4 LTS (x86_64) GCC version: Could not collect Clang version: Could not collect CMake version: version 3.26.4 Libc version: glibc-2.35

Python version: 3.10.14 (main, May 6 2024, 19:42:50) [GCC 11.2.0] (64-bit runtime) Python platform: Linux-5.15.0-97-generic-x86_64-with-glibc2.35 Is CUDA available: True CUDA runtime version: Could not collect CUDA_MODULE_LOADING set to: LAZY GPU models and configuration: GPU 0: NVIDIA GeForce RTX 4090 GPU 1: NVIDIA GeForce RTX 4090

Nvidia driver version: 535.161.07 cuDNN version: Could not collect 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): 32 On-line CPU(s) list: 0-31 Vendor ID: GenuineIntel Model name: Intel(R) Core(TM) i9-14900K CPU family: 6 Model: 183 Thread(s) per core: 2 Core(s) per socket: 24 Socket(s): 1 Stepping: 1 CPU max MHz: 6000.0000 CPU min MHz: 800.0000 BogoMIPS: 6374.40 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid rdseed adx smap clflushopt clwb intel_pt sha_ni xsaveopt xsavec xgetbv1 xsaves split_lock_detect avx_vnni dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp hwp_pkg_req umip pku ospke waitpkg gfni vaes vpclmulqdq tme rdpid movdiri movdir64b fsrm md_clear serialize pconfig arch_lbr flush_l1d arch_capabilities Virtualization: VT-x L1d cache: 896 KiB (24 instances) L1i cache: 1.3 MiB (24 instances) L2 cache: 32 MiB (12 instances) L3 cache: 36 MiB (1 instance) NUMA node(s): 1 NUMA node0 CPU(s): 0-31 Vulnerability Gather data sampling: Not affected Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Not affected Vulnerability Retbleed: Not affected Vulnerability Spec rstack overflow: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Enhanced IBRS, IBPB conditional, RSB filling, PBRSB-eIBRS SW sequence Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected

Versions of relevant libraries: [pip3] numpy==1.26.4 [pip3] optree==0.11.0 [pip3] torch==2.3.1 [pip3] torchaudio==2.3.1 [pip3] torchelastic==0.2.2 [pip3] torchvision==0.18.1 [pip3] triton==2.3.1 [conda] blas 1.0 mkl
[conda] ffmpeg 4.3 hf484d3e_0 pytorch [conda] libjpeg-turbo 2.0.0 h9bf148f_0 pytorch [conda] mkl 2023.1.0 h213fc3f_46344
[conda] mkl-service 2.4.0 py310h5eee18b_1
[conda] mkl_fft 1.3.8 py310h5eee18b_0
[conda] mkl_random 1.2.4 py310hdb19cb5_0
[conda] numpy 1.26.4 py310h5f9d8c6_0
[conda] numpy-base 1.26.4 py310hb5e798b_0
[conda] optree 0.11.0 pypi_0 pypi [conda] pytorch 2.3.1 py3.10_cuda11.8_cudnn8.7.0_0 pytorch [conda] pytorch-cuda 11.8 h7e8668a_5 pytorch [conda] pytorch-mutex 1.0 cuda pytorch [conda] torchaudio 2.3.1 py310_cu118 pytorch [conda] torchelastic 0.2.2 pypi_0 pypi [conda] torchtriton 2.3.1 py310 pytorch [conda] torchvision 0.18.1 py310_cu118 pytorch

pytorch / vision

Inconsistent Behavior with transforms.v2 for Multiple Arguments #8573

🐛 Describe the bug

Versions