mlx
mlx copied to clipboard
ml-explore
MLX: An array framework for Apple silicon
## Proposed changes Adds inner / outer op ## Checklist Put an `x` in the boxes that apply. - [x] I have read the [CONTRIBUTING](https://github.com/ml-explore/mlx/blob/main/CONTRIBUTING.md) document - [x] I have...
```python >>> mx.array([1], dtype=mx.bool_) & mx.array([0], dtype=mx.bool_) TypeError: unsupported operand type(s) for &: 'mlx.core.array' and 'mlx.core.array' >>> mx.logical_and(mx.array([1], dtype=mx.bool_), mx.array([0], dtype=mx.bool_)) AttributeError: module 'mlx.core' has no attribute 'logical_and'. Did you...
To reuse a module copy, like: ```python from copy import deepcopy import mlx.nn as mlx_nn repeat = lambda module, n: [deepcopy(module) for _ in range(n)] mlx_layer = mlx_nn.Linear(2, 3) mlx_layers...
**Describe the bug** - `mlx.core.newaxis` is missing: ```python >>> mx.array([1])[mx.newaxis] AttributeError: module 'mlx.core' has no attribute 'newaxis' ``` - Cannot index with `Ellipsis` / `...`: ```python >>> mx.array([1])[...] ValueError: Cannot...
## Proposed changes Please include a description of the problem or feature this PR is addressing. If there is a corresponding issue, include the issue #326. A formatter structure in...
Hello, I'm trying to train a simple network (mobilenet classifier) which seems fine but I'm getting a segfault after a few batches. Hoping maybe someone can point out what I'm...
## Proposed changes Make array conform to the [Buffer Protocol](https://docs.python.org/3/c-api/buffer.html). The method [`__array__`](https://numpy.org/doc/stable/reference/generated/numpy.ndarray.__array__.html) is replaced by implementing the Python Buffer Protocol. Summary: - The complex implementation of `__array__` is removed....
I think that would be nice for embed devices
Are Swift bindings to MLX in the roadmap/within scope? I attempted to wrap up the built library + metallib file into a macOS `.framework` bundle, giving it a correct `.modulemap`...
## Proposed changes Added ``MaxPool`` and ``AvgPool`` layers. MaxPool and AvgPool layers were requested in [this issue](https://github.com/ml-explore/mlx/issues/25). In this PR, I propose implementing the requested pooling operations by firstly computing...
