diff --git a/declearn/dataset/_split_data.py b/declearn/dataset/_split_data.py index 017c9c08d0a3866f99573ae7b9bf02dfc8837473..1729c2ddea23edb0719a9181c0d2e66d94cab5e6 100644 --- a/declearn/dataset/_split_data.py +++ b/declearn/dataset/_split_data.py @@ -144,7 +144,8 @@ def split_data( ) -> None: """Randomly split a dataset into shards. - The resulting folder structure is : + The resulting folder structure is: + folder/ └─── data*/ └─── client*/ diff --git a/declearn/model/haiku/_vector.py b/declearn/model/haiku/_vector.py index e1f106f141848d5686297c23289d610e025ec1a2..7797ee62c8f09d1d47a52280d847dc870865989a 100644 --- a/declearn/model/haiku/_vector.py +++ b/declearn/model/haiku/_vector.py @@ -53,10 +53,10 @@ class JaxNumpyVector(Vector): Notes ----- - A `JaxnumpyVector` can be operated with either a: - - scalar value - - `NumpyVector` that has similar specifications - - `JaxNumpyVector` that has similar specifications - => resulting in a `JaxNumpyVector` in each of these cases. + - scalar value + - `NumpyVector` that has similar specifications + - `JaxNumpyVector` that has similar specifications + - => resulting in a `JaxNumpyVector` in each of these cases. - The wrapped arrays may be placed on any device (CPU, GPU...) and may not be all on the same device. - The device-placement of the initial `JaxNumpyVector`'s data diff --git a/declearn/model/tensorflow/_vector.py b/declearn/model/tensorflow/_vector.py index 401bbb0665ae95b80a78f0bbb55fa90feb16a3af..079aa17016af9f0947c85d9eb8130c33ca94d914 100644 --- a/declearn/model/tensorflow/_vector.py +++ b/declearn/model/tensorflow/_vector.py @@ -96,10 +96,10 @@ class TensorflowVector(Vector): Notes ----- - A `TensorflowVector` can be operated with either a: - - scalar value - - `NumpyVector` that has similar specifications - - `TensorflowVector` that has similar specifications - => resulting in a `TensorflowVector` in each of these cases. + - scalar value + - `NumpyVector` that has similar specifications + - `TensorflowVector` that has similar specifications + - => resulting in a `TensorflowVector` in each of these cases. - The wrapped tensors may be placed on any device (CPU, GPU...) and may not be all on the same device. - The device-placement of the initial `TensorflowVector`'s data diff --git a/declearn/model/torch/_vector.py b/declearn/model/torch/_vector.py index 662aaa100fde401335d9f80657f32b505986614e..91d93be33bb24ce58592732876f549b91a880b4c 100644 --- a/declearn/model/torch/_vector.py +++ b/declearn/model/torch/_vector.py @@ -48,10 +48,10 @@ class TorchVector(Vector): Notes ----- - A `TorchVector` can be operated with either a: - - scalar value - - `NumpyVector` that has similar specifications - - `TorchVector` that has similar specifications - => resulting in a `TorchVector` in each of these cases. + - scalar value + - `NumpyVector` that has similar specifications + - `TorchVector` that has similar specifications + - => resulting in a `TorchVector` in each of these cases. - The wrapped tensors may be placed on any device (CPU, GPU...) and may not be all on the same device. - The device-placement of the initial `TorchVector`'s data diff --git a/docs/quickstart.md b/docs/quickstart.md index be711050f18ecd067b2898ade1bdca86d082b035..f2c4d8f681d06434a4046c1bc630858e3920a076 100644 --- a/docs/quickstart.md +++ b/docs/quickstart.md @@ -52,8 +52,8 @@ declearn-quickrun --config "examples/mnist_quickrun/config.toml" **To better understand the details** of what happens under the hood you can look at what the key element of the declearn process are in -[section 1.2.](#12-python-script). To understand how to use the quickrun mode -in practice, see [section 2.1.](#21-quickrun-on-your-problem). +[section 1.2](#12-python-script). To understand how to use the quickrun mode +in practice, see [section 2.1](#21-quickrun-on-your-problem). ### 1.2. Python script