b8591340

There’s seems to be no way to modify the `_elements` memory in place when we want to alter a property that does not partake in the `Equatable` and `Hashable` conformance....

`AssertionError` in `min` op with `len(inputs) == 2` for `sentence-transformers/all-mpnet-base-v2`

1

```python import numpy as np import torch from transformers import AutoTokenizer, AutoModel import coremltools as ct name = "sentence-transformers/all-mpnet-base-v2" length = 384 sentence = 'This is an example sentence' tokenizer...

Error converting CLIP PyTorch model to CoreML: convert function for op `baddbmm` not implemented

3

``` pip install coremltools --pre -U pip install git+https://github.com/openai/CLIP.git ``` ```python import torch import clip import coremltools as ct import numpy as np from PIL import Image device="cpu" model, preprocess...

`#[derive(Archive)]` with `#[repr(packed)]` needs to conform to `unaligned_references`

6

The traditional behavior is UB and will become an error as per https://github.com/rust-lang/rust/issues/82523.

Failure at `assert "PackedParams" in node.output().type().name()` when converting Silero VAD

1

```python import torch import coremltools as ct import numpy as np model, _ = torch.hub.load( repo_or_dir="snakers4/silero-vad", model="silero_vad", force_reload=True ) model.eval() seconds_of_audio = 60 sample_rate = 16000 num_samples = sample_rate *...