[0.3.14] TensorNeko Release

Fix:

• Fix ar_1d AR calculation when the num proposals is less than the max_proposals.

[0.3.13] TensorNeko Release

Features:

• Add file_key arguments for ap_1d and ar_1d
• Add stub for tensorneko_lib

Fix:

• Wrong computation of ap_1d

[0.3.12] TensorNeko Release

Features:

• new library tensorneko_lib for high performance rust implementations.
• npy reader and writer in tensorneko_util.io
• tensorneko_util.Timer.lines to automatically time each lines in the wrapped function.

Fix:

• opencv image writer cannot save tensor as image.

[0.3.11] TensorNeko Release

Features:

• Singleton support with class with init args
• More utils for download files
• Typing for Registry

[0.3.10] TensorNeko Release

Features:

• Registry for easier decorator-style class/func registration
• read and write are also in the top level of tensorneko_util

[0.3.9] TensorNeko Release

Features:

• JsonReader now can load jsonl files
• WeightReader and WeightWriter that can read and write the pt, ckpt, safetensors files
• read and write now is in the top level of the tensorneko package
• import_tqdm and import_tqdm_auto now are not private

[0.3.8] TensorNeko Release

Features:

• Group the metrics, losses, and system stats in logger
• Add input_shape arg for BinaryClassifier

Fix:

• NekoModel cannot train when the logger is None

[0.3.7] TensorNeko Release

Fix:

• Training error when using lightning 2.1
• Docs are not clear for VideoReader, RoundRobinDataset, NestedDataset, ListDataset.

[0.3.6] TensorNeko Release

Features:

• Better type hint for eventbus and dispatch
• Dynamic import for mita_tqdm
• on_step and on_epoch arguments for GpuStatsLogger and SystemStatsLogger
• Timer now doesn't require parentheses as decorator

Fix:

• crop_with_padding using numpy array
• sample_indexes not imported
• FID not working with progress_bar=True
• import error with pytorch lightning 2.1.3

[0.3.5] TensorNeko Release

Features:

• Add support for pytorch 2.1 and lightning 2.1
• Auto infer for notebook.display, now supports audio, video, image, and code.
• ListDataset, like TensorDataset in torch but directly uses list
• SequentialIterSampler, can split dataset to multiple epochs. Normally can be used for training in large-scale dataset but want to utilize the learning rate scheduler based on epochs.
• BinaryClassifier. A predefined model for binary classification.
• sample_indexes. Sample a continious indexes giving a source sequence length.

Fix:

• Import problem raise by RoundRobinDataset
• FID calculation problem when set num_workers > 0.
• iou_2d calculation error