Eaxmple in readme refers to function that doesn't appear to exist

[owainkenwayucl]

In the Readme.md there is this example for eager mode:

import torch
import torch_ttnn

model = YourModel()

device = ttnn.open_device(device_id=0)
model.to(torch_ttnn.ttnn_device_as_torch_device(device))

output = model(input_data)

If you actually try to do this with a model you discover that torch_ttnn.tnn_device_as_torch_device() doesn't seem to exist.

(pytorch20-ttnn) ❄️  (uccaoke) venvs≫ python3
Python 3.10.12 (main, Nov  4 2025, 08:48:33) [GCC 11.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
>>> import ttnn
2025-11-29 22:18:13.054 | DEBUG    | ttnn:<module>:77 - Initial ttnn.CONFIG:
Config{cache_path=/home/uccaoke/.cache/ttnn,model_cache_path=/home/uccaoke/.cache/ttnn/models,tmp_dir=/tmp/ttnn,enable_model_cache=false,enable_fast_runtime_mode=true,throw_exception_on_fallback=false,enable_logging=false,enable_graph_report=false,enable_detailed_buffer_report=false,enable_detailed_tensor_report=false,enable_comparison_mode=false,comparison_mode_should_raise_exception=false,comparison_mode_pcc=0.9999,root_report_path=generated/ttnn/reports,report_name=std::nullopt,std::nullopt}
>>> import torch_ttnn
>>> device = ttnn.open_device(device_id=0)

( cut because way way too much output )

2025-11-29 22:19:53.752 | info     |             UMD | Mapped hugepage 0x2180000000 to NOC address 0x880000000 (sysmem_manager.cpp:247)
2025-11-29 22:19:53.752 | info     |             UMD | Pinning pages for Hugepage: virtual address 0x7f2480000000 and size 0x30000000 pinned to physical address 0x2140000000 and mapped to noc address 0x8c0000000 (pci_device.cpp:644)
2025-11-29 22:19:53.752 | info     |             UMD | Mapped hugepage 0x2140000000 to NOC address 0x8c0000000 (sysmem_manager.cpp:247)
2025-11-29 22:19:53.825 | info     |          Fabric | TopologyMapper mapping start (mesh=0): n_log=32, n_phys=32, log_deg_hist={2:4, 3:16, 4:12}, phys_deg_hist={4:32} (topology_mapper.cpp:574)
2025-11-29 22:19:54.112 | warning  |           Metal | Opening subset of mmio devices slows down UMD read/write to remote chips. If opening more devices, consider using CreateDevices API. (device_pool.cpp:303)
2025-11-29 22:19:54.113 | info     |           Metal | Profiler started on device 0 (device_pool.cpp:203)
>>> d = torch_ttnn.ttnn_device_as_torch_device(device)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
AttributeError: module 'torch_ttnn' has no attribute 'ttnn_device_as_torch_device
>>>

I've done some poking through the code-base but I can't really see what this is meant to be...