# Yoyodyne 🪀
[](https://pypi.org/project/yoyodyne)
[](https://pypi.org/project/yoyodyne)
[](https://circleci.com/gh/CUNY-CL/yoyodyne/tree/master)
Yoyodyne provides neural models for small-vocabulary sequence-to-sequence
generation with and without feature conditioning.
These models are implemented using [PyTorch](https://pytorch.org/) and
[Lightning](https://www.pytorchlightning.ai/).
While we provide classic LSTM and transformer models, some of the provided
models are particularly well-suited for problems where the source-target
alignments are roughly monotonic (e.g., `transducer`) and/or where source and
target vocabularies have substantial overlap (e.g., `pointer_generator_lstm`).
## Philosophy
Yoyodyne is inspired by [FairSeq](https://github.com/facebookresearch/fairseq)
(Ott et al. 2019) but differs on several key points of design:
- It is for small-vocabulary sequence-to-sequence generation, and therefore
includes no affordances for machine translation or language modeling.
Because of this:
- It has no plugin interface and the architectures provided are intended
to be reasonably exhaustive.
- There is little need for data preprocessing; it works with TSV files.
- It has support for using features to condition decoding, with
architecture-specific code to handle feature information.
- It uses validation accuracy (not loss) for model selection and early
stopping.
- Releases are made regularly.
- 🚧 UNDER CONSTRUCTION 🚧: It has exhaustive test suites.
- 🚧 UNDER CONSTRUCTION 🚧: It has performance benchmarks.
## Installation
### Local installation
Yoyodyne currently supports Python 3.9 and 3.10.
[#60](https://github.com/CUNY-CL/yoyodyne/issues/60) is a known blocker to
Python \> 3.10 support.
First install dependencies:
pip install -r requirements.txt
Then install:
pip install .
It can then be imported like a regular Python module:
```python
import yoyodyne
```
### Google Colab
Yoyodyne is compatible with [Google Colab](https://colab.research.google.com/)
GPU runtimes. [This
notebook](https://colab.research.google.com/drive/1O4VWvpqLrCxxUvyYMbGH9HOyXQSoh5bP?usp=sharing)
provides a worked example. Colab also provides access to TPU runtimes, but this
is not yet compatible with Yoyodyne to our knowledge.
## Usage
### Training
Training is performed by the [`yoyodyne-train`](yoyodyne/train.py) script. One
must specify the following required arguments:
- `--model_dir`: path for model metadata and checkpoints
- `--experiment`: name of experiment (pick something unique)
- `--train`: path to TSV file containing training data
- `--val`: path to TSV file containing validation data
The user can also specify various optional training and architectural arguments.
See below or run [`yoyodyne-train --help`](yoyodyne/train.py) for more
information.
### Prediction
Prediction is performed by the [`yoyodyne-predict`](yoyodyne/predict.py) script.
One must specify the following required arguments:
- `--model_dir`: path for model metadata
- `--experiment`: name of experiment
- `--checkpoint`: path to checkpoint
- `--predict`: path to file containing data to be predicted
- `--output`: path for predictions
The `--predict` file can either be a TSV file or an ordinary TXT file with one
source string per line; in the latter case, specify `--target_col 0`. Run
[`yoyodyne-predict --help`](yoyodyne/predict.py) for more information.
## Data format
The default data format is a two-column TSV file in which the first column is
the source string and the second the target string.
source target
To enable the use of a feature column, one specifies a (non-zero) argument to
`--features_col`. For instance in the [SIGMORPHON 2017 shared
task](https://sigmorphon.github.io/sharedtasks/2017/), the first column is the
source (a lemma), the second is the target (the inflection), and the third
contains semi-colon delimited feature strings:
source target feat1;feat2;...
this format is specified by `--features_col 3`.
Alternatively, for the [SIGMORPHON 2016 shared
task](https://sigmorphon.github.io/sharedtasks/2016/) data:
source feat1,feat2,... target
this format is specified by `--features_col 2 --features_sep , --target_col 3`.
In order to ensure that targets are ignored during prediction, one can specify
`--target_col 0`.
## Reserved symbols
Yoyodyne reserves symbols of the form `<...>` for internal use.
Feature-conditioned models also use `[...]` to avoid clashes between feature
symbols and source and target symbols. Therefore, users should not provide any
symbols of the form `<...>` or `[...]`.
## Model checkpointing
Checkpointing is handled by
[Lightning](https://pytorch-lightning.readthedocs.io/en/stable/common/checkpointing_basic.html).
The path for model information, including checkpoints, is specified by a
combination of `--model_dir` and `--experiment`, such that we build the path
`model_dir/experiment/version_n`, where each run of an experiment with the same
`model_dir` and `experiment` is namespaced with a new version number. A version
stores everything needed to reload the model, including the hyperparameters
(`model_dir/experiment_name/version_n/hparams.yaml`) and the checkpoints
directory (`model_dir/experiment_name/version_n/checkpoints`).
By default, each run initializes a new model from scratch, unless the
`--train_from` argument is specified. To continue training from a specific
checkpoint, the **full path to the checkpoint** should be specified with for the
`--train_from` argument. This creates a new version, but starts training from
the provided model checkpoint.
During training, we save the best `--save_top_k` checkpoints (by default, 1)
ranked according to accuracy on the `--val` set. For example, `--save_top_k 5`
will save the top 5 most accurate models.
## Models
The user specifies the overall architecture for the model using the `--arch`
flag. The value of this flag specifies the decoder's architecture and whether or
not an attention mechanism is present. This flag also specifies a default
architecture for the encoder(s), but it is possible to override this with
additional flags. Supported values for `--arch` are:
- `attentive_lstm`: This is an LSTM decoder with LSTM encoders (by default)
and an attention mechanism. The initial hidden state is treated as a learned
parameter.
- `lstm`: This is an LSTM decoder with LSTM encoders (by default); in lieu of
an attention mechanism, the last non-padding hidden state of the encoder is
concatenated with the decoder hidden state.
- `pointer_generator_lstm`: This is an LSTM decoder with LSTM encoders (by
default) and a pointer-generator mechanism. Since this model contains a copy
mechanism, it may be superior to an ordinary attentive LSTM when the source
and target vocabularies overlap significantly. Note that this model requires
that the number of `--encoder_layers` and `--decoder_layers` match.
- `pointer_generator_transformer`: This is a transformer decoder with
transformer encoders (by default) and a pointer-generator mechanism. Like
`pointer_generator_lstm`, it may be superior to an ordinary transformer when
the source and target vocabularies overlap significantly. When using
features, the user may wish to specify the n umber of features attention
heads (with `--features_attention_heads`; default: `1`).
- `transducer`: This is an LSTM decoder with LSTM encoders (by default) and a
neural transducer mechanism. On model creation, expectation maximization is
used to learn a sequence of edit operations, and imitation learning is used
to train the model to implement the oracle policy, with roll-in controlled
by the `--oracle_factor` flag (default: `1`). Since this model assumes
monotonic alignment, it may be superior to attentive models when the
alignment between input and output is roughly monotonic and when input and
output vocabularies overlap significantly.
- `transformer`: This is a transformer decoder with transformer encoders (by
default). Sinusodial positional encodings and layer normalization are used.
The user may wish to specify the number of attention heads (with
`--source_attention_heads`; default: `4`).
The user can override the default encoder architectures. One can override the
source encoder using the `--source_encoder` flag:
- `feature_invariant_transformer`: This is a variant of the transformer
encoder used with features; it concatenates source and features and uses a
learned embedding to distinguish between source and features symbols.
- `linear`: This is a linear encoder.
- `lstm`: This is a LSTM encoder.
- `transformer`: This is a transformer encoder.
When using features, the user can also specify a non-default features encoder
using the `--features_encoder` flag (`linear`, `lstm`, `transformer`).
For all models, the user may also wish to specify:
- `--decoder_layers` (default: `1`): number of decoder layers
- `--embedding` (default: `128`): embedding size
- `--encoder_layers` (default: `1`): number of encoder layers
- `--hidden_size` (default: `512`): hidden layer size
By default, LSTM encoders are bidirectional. One can disable this with the
`--no_bidirectional` flag.
## Training options
A non-exhaustive list includes:
- Batch size:
- `--batch_size` (default: `32`)
- `--accumulate_grad_batches` (default: not enabled)
- Regularization:
- `--dropout` (default: `0.2`)
- `--label_smoothing` (default: `0.0`)
- `--gradient_clip_val` (default: not enabled)
- Optimizer:
- `--learning_rate` (default: `0.001`)
- `--optimizer` (default: `"adam"`)
- `--beta1` (default: `0.9`): $\beta_1$ hyperparameter for the Adam
optimizer (`--optimizer adam`)
- `--beta2` (default: `0.99`): $\beta_2$ hyperparameter for the Adam
optimizer (`--optimizer adam`)
- `--scheduler` (default: not enabled)
- Duration:
- `--max_epochs`
- `--min_epochs`
- `--max_steps`
- `--min_steps`
- `--max_time`
- `--patience`
- Seeding:
- `--seed`
- [Weights & Biases](https://wandb.ai/site):
- `--log_wandb` (default: `False`): enables Weights & Biases tracking
### Schedulers
By default, Yoyodyne uses a constant learning rate during training, but best
practice is to gradually decreasing learning rate as the model approaches
convergence using a [scheduler](yoyodyne/schedulers.py). Three (non-null)
schedulers are supported and are selected with `--scheduler`:
- `lineardecay`: linearly decreases the learning rate (multiplying it by
`--start_factor`) for `--total_decay_steps` steps, then decreases the
learning rate by `--end_factor`.
- `reduceonplateau`: reduces the learning rate (multiplying it by
`--reduceonplateau_factor`) after `--reduceonplateau_patience` epochs with
no improvement (when the loss stops decreasing if `--reduceonplateau loss`,
or when the validation accuracy stops increasing if
`--reduceonplateaumode accuracy`) until the learning rate is less than or
equal to `--min_learning_rate`.
- `warmupinvsqrt`: linearly increases the learning rate from 0 to
`--learning_rate` for `--warmup_steps` steps, then decreases learning rate
according to an inverse root square schedule.
### Simulating large batches
At times one may wish to train with a larger batch size than will fit in "in
core". For example, suppose one wishes to fit with a batch size of 4,096, but
this gives an out of memory exception. Then, with minimal overhead, one could
simulate an effective batch size of 4,096 by using batches (`--batch_size`) of
1,024, accumulating gradients from 4 batches
([`--accumulate_grad_batches`](https://lightning.ai/docs/pytorch/stable/common/optimization.html#id3))
per update:
yoyodyne-train --batch_size 1024 --accumulate_grad_batches 4 ...
### Automatic tuning
`yododyne-train --auto_lr_find` uses a heuristic ([Smith
2017](https://ieeexplore.ieee.org/abstract/document/7926641)) to propose an
initial learning rate. Batch auto-scaling is not supported.
### Hyperparameter tuning
**No neural model should be deployed without proper hyperparameter tuning.**
However, the default options give a reasonable initial settings for an attentive
biLSTM. For transformer-based architectures, experiment with multiple encoder
and decoder layers, much larger batches, and the warmup-plus-inverse square root
decay scheduler.
### Weights & Biases tuning
[`wandb_sweeps`](examples/wandb_sweeps) shows how to use [Weights &
Biases](https://wandb.ai/site) to run hyperparameter sweeps.
## Accelerators
[Hardware
accelerators](https://pytorch-lightning.readthedocs.io/en/stable/extensions/accelerator.html)
can be used during training or prediction. In addition to CPU (the default) and
GPU (`--accelerator gpu`), [other
accelerators](https://pytorch-lightning.readthedocs.io/en/stable/extensions/accelerator.html)
may also be supported but not all have been tested yet.
## Precision
By default, training uses 32-bit precision. However, the `--precision` flag
allows the user to perform training with half precision (`16`) or with the
[`bfloat16` half precision
format](https://en.wikipedia.org/wiki/Bfloat16_floating-point_format) if
supported by the accelerator. This may reduce the size of the model and batches
in memory, allowing one to use larger batches.
## Examples
The [`examples`](examples) directory contains interesting examples, including:
- [`wandb_sweeps`](examples/wandb_sweeps) shows how to use [Weights &
Biases](https://wandb.ai/site) to run hyperparameter sweeps.
## For developers
*Developers, developers, developers!* - Steve Ballmer
This section contains instructions for the Yoyodyne maintainers.
### Releasing
1. Create a new branch. E.g., if you want to call this branch "release":
`git checkout -b release`
2. Sync your fork's branch to the upstream master branch. E.g., if the upstream
remote is called "upstream": `git pull upstream master`
3. Increment the version field in [`pyproject.toml`](pyproject.toml).
4. Stage your changes: `git add pyproject.toml`.
5. Commit your changes: `git commit -m "your commit message here"`
6. Push your changes. E.g., if your branch is called "release":
`git push origin release`
7. Submit a PR for your release and wait for it to be merged into `master`.
8. Tag the `master` branch's last commit. The tag should begin with `v`; e.g.,
if the new version is 3.1.4, the tag should be `v3.1.4`. This can be done:
- on GitHub itself: click the "Releases" or "Create a new release" link on
the right-hand side of the Yoyodyne GitHub page) and follow the
dialogues.
- from the command-line using `git tag`.
9. Build the new release: `python -m build`
10. Upload the result to PyPI: `twine upload dist/*`
## References
Ott, M., Edunov, S., Baevski, A., Fan, A., Gross, S., Ng, N., Grangier, D., and
Auli, M. 2019. [fairseq: a fast, extensible toolkit for sequence
modeling](https://aclanthology.org/N19-4009/). In *Proceedings of the 2019
Conference of the North American Chapter of the Association for Computational
Linguistics (Demonstrations)*, pages 48-53.
Smith, L. N. 2017. [Cyclical learning rates for training neural
networks](https://ieeexplore.ieee.org/abstract/document/7926641). In *2017 IEEE
Winter Conference on Applications of Computer Vision*, pages 464-472.
Raw data
{
"_id": null,
"home_page": "",
"name": "yoyodyne",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": "",
"keywords": "computational linguistics,morphology,natural language processing,language",
"author": "Adam Wiemerslage, Kyle Gorman, Travis Bartley",
"author_email": "",
"download_url": "https://files.pythonhosted.org/packages/65/0b/846e0e8ceaf43e5fc212f50bf9e5a97fbcd3c2b71525d0ba7cf4610b333c/yoyodyne-0.2.9.tar.gz",
"platform": null,
"description": "# Yoyodyne \ud83e\ude80\n\n[](https://pypi.org/project/yoyodyne)\n[](https://pypi.org/project/yoyodyne)\n[](https://circleci.com/gh/CUNY-CL/yoyodyne/tree/master)\n\nYoyodyne provides neural models for small-vocabulary sequence-to-sequence\ngeneration with and without feature conditioning.\n\nThese models are implemented using [PyTorch](https://pytorch.org/) and\n[Lightning](https://www.pytorchlightning.ai/).\n\nWhile we provide classic LSTM and transformer models, some of the provided\nmodels are particularly well-suited for problems where the source-target\nalignments are roughly monotonic (e.g., `transducer`) and/or where source and\ntarget vocabularies have substantial overlap (e.g., `pointer_generator_lstm`).\n\n## Philosophy\n\nYoyodyne is inspired by [FairSeq](https://github.com/facebookresearch/fairseq)\n(Ott et al.\u00a02019) but differs on several key points of design:\n\n- It is for small-vocabulary sequence-to-sequence generation, and therefore\n includes no affordances for machine translation or language modeling.\n Because of this:\n - It has no plugin interface and the architectures provided are intended\n to be reasonably exhaustive.\n - There is little need for data preprocessing; it works with TSV files.\n- It has support for using features to condition decoding, with\n architecture-specific code to handle feature information.\n- It uses validation accuracy (not loss) for model selection and early\n stopping.\n- Releases are made regularly.\n- \ud83d\udea7 UNDER CONSTRUCTION \ud83d\udea7: It has exhaustive test suites.\n- \ud83d\udea7 UNDER CONSTRUCTION \ud83d\udea7: It has performance benchmarks.\n\n## Installation\n\n### Local installation\n\nYoyodyne currently supports Python 3.9 and 3.10.\n[#60](https://github.com/CUNY-CL/yoyodyne/issues/60) is a known blocker to\nPython \\> 3.10 support.\n\nFirst install dependencies:\n\n pip install -r requirements.txt\n\nThen install:\n\n pip install .\n\nIt can then be imported like a regular Python module:\n\n```python\nimport yoyodyne\n```\n\n### Google Colab\n\nYoyodyne is compatible with [Google Colab](https://colab.research.google.com/)\nGPU runtimes. [This\nnotebook](https://colab.research.google.com/drive/1O4VWvpqLrCxxUvyYMbGH9HOyXQSoh5bP?usp=sharing)\nprovides a worked example. Colab also provides access to TPU runtimes, but this\nis not yet compatible with Yoyodyne to our knowledge.\n\n## Usage\n\n### Training\n\nTraining is performed by the [`yoyodyne-train`](yoyodyne/train.py) script. One\nmust specify the following required arguments:\n\n- `--model_dir`: path for model metadata and checkpoints\n- `--experiment`: name of experiment (pick something unique)\n- `--train`: path to TSV file containing training data\n- `--val`: path to TSV file containing validation data\n\nThe user can also specify various optional training and architectural arguments.\nSee below or run [`yoyodyne-train --help`](yoyodyne/train.py) for more\ninformation.\n\n### Prediction\n\nPrediction is performed by the [`yoyodyne-predict`](yoyodyne/predict.py) script.\nOne must specify the following required arguments:\n\n- `--model_dir`: path for model metadata\n- `--experiment`: name of experiment\n- `--checkpoint`: path to checkpoint\n- `--predict`: path to file containing data to be predicted\n- `--output`: path for predictions\n\nThe `--predict` file can either be a TSV file or an ordinary TXT file with one\nsource string per line; in the latter case, specify `--target_col 0`. Run\n[`yoyodyne-predict --help`](yoyodyne/predict.py) for more information.\n\n## Data format\n\nThe default data format is a two-column TSV file in which the first column is\nthe source string and the second the target string.\n\n source target\n\nTo enable the use of a feature column, one specifies a (non-zero) argument to\n`--features_col`. For instance in the [SIGMORPHON 2017 shared\ntask](https://sigmorphon.github.io/sharedtasks/2017/), the first column is the\nsource (a lemma), the second is the target (the inflection), and the third\ncontains semi-colon delimited feature strings:\n\n source target feat1;feat2;...\n\nthis format is specified by `--features_col 3`.\n\nAlternatively, for the [SIGMORPHON 2016 shared\ntask](https://sigmorphon.github.io/sharedtasks/2016/) data:\n\n source feat1,feat2,... target\n\nthis format is specified by `--features_col 2 --features_sep , --target_col 3`.\n\nIn order to ensure that targets are ignored during prediction, one can specify\n`--target_col 0`.\n\n## Reserved symbols\n\nYoyodyne reserves symbols of the form `<...>` for internal use.\nFeature-conditioned models also use `[...]` to avoid clashes between feature\nsymbols and source and target symbols. Therefore, users should not provide any\nsymbols of the form `<...>` or `[...]`.\n\n## Model checkpointing\n\nCheckpointing is handled by\n[Lightning](https://pytorch-lightning.readthedocs.io/en/stable/common/checkpointing_basic.html).\nThe path for model information, including checkpoints, is specified by a\ncombination of `--model_dir` and `--experiment`, such that we build the path\n`model_dir/experiment/version_n`, where each run of an experiment with the same\n`model_dir` and `experiment` is namespaced with a new version number. A version\nstores everything needed to reload the model, including the hyperparameters\n(`model_dir/experiment_name/version_n/hparams.yaml`) and the checkpoints\ndirectory (`model_dir/experiment_name/version_n/checkpoints`).\n\nBy default, each run initializes a new model from scratch, unless the\n`--train_from` argument is specified. To continue training from a specific\ncheckpoint, the **full path to the checkpoint** should be specified with for the\n`--train_from` argument. This creates a new version, but starts training from\nthe provided model checkpoint.\n\nDuring training, we save the best `--save_top_k` checkpoints (by default, 1)\nranked according to accuracy on the `--val` set. For example, `--save_top_k 5`\nwill save the top 5 most accurate models.\n\n## Models\n\nThe user specifies the overall architecture for the model using the `--arch`\nflag. The value of this flag specifies the decoder's architecture and whether or\nnot an attention mechanism is present. This flag also specifies a default\narchitecture for the encoder(s), but it is possible to override this with\nadditional flags. Supported values for `--arch` are:\n\n- `attentive_lstm`: This is an LSTM decoder with LSTM encoders (by default)\n and an attention mechanism. The initial hidden state is treated as a learned\n parameter.\n- `lstm`: This is an LSTM decoder with LSTM encoders (by default); in lieu of\n an attention mechanism, the last non-padding hidden state of the encoder is\n concatenated with the decoder hidden state.\n- `pointer_generator_lstm`: This is an LSTM decoder with LSTM encoders (by\n default) and a pointer-generator mechanism. Since this model contains a copy\n mechanism, it may be superior to an ordinary attentive LSTM when the source\n and target vocabularies overlap significantly. Note that this model requires\n that the number of `--encoder_layers` and `--decoder_layers` match.\n- `pointer_generator_transformer`: This is a transformer decoder with\n transformer encoders (by default) and a pointer-generator mechanism. Like\n `pointer_generator_lstm`, it may be superior to an ordinary transformer when\n the source and target vocabularies overlap significantly. When using\n features, the user may wish to specify the n umber of features attention\n heads (with `--features_attention_heads`; default: `1`).\n- `transducer`: This is an LSTM decoder with LSTM encoders (by default) and a\n neural transducer mechanism. On model creation, expectation maximization is\n used to learn a sequence of edit operations, and imitation learning is used\n to train the model to implement the oracle policy, with roll-in controlled\n by the `--oracle_factor` flag (default: `1`). Since this model assumes\n monotonic alignment, it may be superior to attentive models when the\n alignment between input and output is roughly monotonic and when input and\n output vocabularies overlap significantly.\n- `transformer`: This is a transformer decoder with transformer encoders (by\n default). Sinusodial positional encodings and layer normalization are used.\n The user may wish to specify the number of attention heads (with\n `--source_attention_heads`; default: `4`).\n\nThe user can override the default encoder architectures. One can override the\nsource encoder using the `--source_encoder` flag:\n\n- `feature_invariant_transformer`: This is a variant of the transformer\n encoder used with features; it concatenates source and features and uses a\n learned embedding to distinguish between source and features symbols.\n- `linear`: This is a linear encoder.\n- `lstm`: This is a LSTM encoder.\n- `transformer`: This is a transformer encoder.\n\nWhen using features, the user can also specify a non-default features encoder\nusing the `--features_encoder` flag (`linear`, `lstm`, `transformer`).\n\nFor all models, the user may also wish to specify:\n\n- `--decoder_layers` (default: `1`): number of decoder layers\n- `--embedding` (default: `128`): embedding size\n- `--encoder_layers` (default: `1`): number of encoder layers\n- `--hidden_size` (default: `512`): hidden layer size\n\nBy default, LSTM encoders are bidirectional. One can disable this with the\n`--no_bidirectional` flag.\n\n## Training options\n\nA non-exhaustive list includes:\n\n- Batch size:\n - `--batch_size` (default: `32`)\n - `--accumulate_grad_batches` (default: not enabled)\n- Regularization:\n - `--dropout` (default: `0.2`)\n - `--label_smoothing` (default: `0.0`)\n - `--gradient_clip_val` (default: not enabled)\n- Optimizer:\n - `--learning_rate` (default: `0.001`)\n - `--optimizer` (default: `\"adam\"`)\n - `--beta1` (default: `0.9`): $\\beta_1$ hyperparameter for the Adam\n optimizer (`--optimizer adam`)\n - `--beta2` (default: `0.99`): $\\beta_2$ hyperparameter for the Adam\n optimizer (`--optimizer adam`)\n - `--scheduler` (default: not enabled)\n- Duration:\n - `--max_epochs`\n - `--min_epochs`\n - `--max_steps`\n - `--min_steps`\n - `--max_time`\n - `--patience`\n- Seeding:\n - `--seed`\n- [Weights & Biases](https://wandb.ai/site):\n - `--log_wandb` (default: `False`): enables Weights & Biases tracking\n\n### Schedulers\n\nBy default, Yoyodyne uses a constant learning rate during training, but best\npractice is to gradually decreasing learning rate as the model approaches\nconvergence using a [scheduler](yoyodyne/schedulers.py). Three (non-null)\nschedulers are supported and are selected with `--scheduler`:\n\n- `lineardecay`: linearly decreases the learning rate (multiplying it by\n `--start_factor`) for `--total_decay_steps` steps, then decreases the\n learning rate by `--end_factor`.\n- `reduceonplateau`: reduces the learning rate (multiplying it by\n `--reduceonplateau_factor`) after `--reduceonplateau_patience` epochs with\n no improvement (when the loss stops decreasing if `--reduceonplateau loss`,\n or when the validation accuracy stops increasing if\n `--reduceonplateaumode accuracy`) until the learning rate is less than or\n equal to `--min_learning_rate`.\n- `warmupinvsqrt`: linearly increases the learning rate from 0 to\n `--learning_rate` for `--warmup_steps` steps, then decreases learning rate\n according to an inverse root square schedule.\n\n### Simulating large batches\n\nAt times one may wish to train with a larger batch size than will fit in \"in\ncore\". For example, suppose one wishes to fit with a batch size of 4,096, but\nthis gives an out of memory exception. Then, with minimal overhead, one could\nsimulate an effective batch size of 4,096 by using batches (`--batch_size`) of\n1,024, accumulating gradients from 4 batches\n([`--accumulate_grad_batches`](https://lightning.ai/docs/pytorch/stable/common/optimization.html#id3))\nper update:\n\n yoyodyne-train --batch_size 1024 --accumulate_grad_batches 4 ...\n\n### Automatic tuning\n\n`yododyne-train --auto_lr_find` uses a heuristic ([Smith\n2017](https://ieeexplore.ieee.org/abstract/document/7926641)) to propose an\ninitial learning rate. Batch auto-scaling is not supported.\n\n### Hyperparameter tuning\n\n**No neural model should be deployed without proper hyperparameter tuning.**\nHowever, the default options give a reasonable initial settings for an attentive\nbiLSTM. For transformer-based architectures, experiment with multiple encoder\nand decoder layers, much larger batches, and the warmup-plus-inverse square root\ndecay scheduler.\n\n### Weights & Biases tuning\n\n[`wandb_sweeps`](examples/wandb_sweeps) shows how to use [Weights &\nBiases](https://wandb.ai/site) to run hyperparameter sweeps.\n\n## Accelerators\n\n[Hardware\naccelerators](https://pytorch-lightning.readthedocs.io/en/stable/extensions/accelerator.html)\ncan be used during training or prediction. In addition to CPU (the default) and\nGPU (`--accelerator gpu`), [other\naccelerators](https://pytorch-lightning.readthedocs.io/en/stable/extensions/accelerator.html)\nmay also be supported but not all have been tested yet.\n\n## Precision\n\nBy default, training uses 32-bit precision. However, the `--precision` flag\nallows the user to perform training with half precision (`16`) or with the\n[`bfloat16` half precision\nformat](https://en.wikipedia.org/wiki/Bfloat16_floating-point_format) if\nsupported by the accelerator. This may reduce the size of the model and batches\nin memory, allowing one to use larger batches.\n\n## Examples\n\nThe [`examples`](examples) directory contains interesting examples, including:\n\n- [`wandb_sweeps`](examples/wandb_sweeps) shows how to use [Weights &\n Biases](https://wandb.ai/site) to run hyperparameter sweeps.\n\n## For developers\n\n*Developers, developers, developers!* - Steve Ballmer\n\nThis section contains instructions for the Yoyodyne maintainers.\n\n### Releasing\n\n1. Create a new branch. E.g., if you want to call this branch \"release\":\n `git checkout -b release`\n2. Sync your fork's branch to the upstream master branch. E.g., if the upstream\n remote is called \"upstream\": `git pull upstream master`\n3. Increment the version field in [`pyproject.toml`](pyproject.toml).\n4. Stage your changes: `git add pyproject.toml`.\n5. Commit your changes: `git commit -m \"your commit message here\"`\n6. Push your changes. E.g., if your branch is called \"release\":\n `git push origin release`\n7. Submit a PR for your release and wait for it to be merged into `master`.\n8. Tag the `master` branch's last commit. The tag should begin with `v`; e.g.,\n if the new version is 3.1.4, the tag should be `v3.1.4`. This can be done:\n - on GitHub itself: click the \"Releases\" or \"Create a new release\" link on\n the right-hand side of the Yoyodyne GitHub page) and follow the\n dialogues.\n - from the command-line using `git tag`.\n9. Build the new release: `python -m build`\n10. Upload the result to PyPI: `twine upload dist/*`\n\n## References\n\nOtt, M., Edunov, S., Baevski, A., Fan, A., Gross, S., Ng, N., Grangier, D., and\nAuli, M. 2019. [fairseq: a fast, extensible toolkit for sequence\nmodeling](https://aclanthology.org/N19-4009/). In *Proceedings of the 2019\nConference of the North American Chapter of the Association for Computational\nLinguistics (Demonstrations)*, pages 48-53.\n\nSmith, L. N. 2017. [Cyclical learning rates for training neural\nnetworks](https://ieeexplore.ieee.org/abstract/document/7926641). In *2017 IEEE\nWinter Conference on Applications of Computer Vision*, pages 464-472.\n",
"bugtrack_url": null,
"license": "Apache 2.0",
"summary": "Small-vocabulary neural sequence-to-sequence models",
"version": "0.2.9",
"project_urls": {
"homepage": "https://github.com/CUNY-CL/yoyodyne"
},
"split_keywords": [
"computational linguistics",
"morphology",
"natural language processing",
"language"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "8e933cce83abdb0212794b6529b2104bb6d0e4057f66d0ea2a8d8c6336165f54",
"md5": "8ed2ca973a7855fe660894518f87aa93",
"sha256": "826acf209c14fc9ff5c323e82b582e878fdc6bc9abbcb2f6a3e11f3ab1b024f3"
},
"downloads": -1,
"filename": "yoyodyne-0.2.9-py3-none-any.whl",
"has_sig": false,
"md5_digest": "8ed2ca973a7855fe660894518f87aa93",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 69774,
"upload_time": "2024-03-06T19:43:58",
"upload_time_iso_8601": "2024-03-06T19:43:58.893381Z",
"url": "https://files.pythonhosted.org/packages/8e/93/3cce83abdb0212794b6529b2104bb6d0e4057f66d0ea2a8d8c6336165f54/yoyodyne-0.2.9-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "650b846e0e8ceaf43e5fc212f50bf9e5a97fbcd3c2b71525d0ba7cf4610b333c",
"md5": "da869c009d74806121a207ab0a2fcce7",
"sha256": "c47deadba2661e92e2984dfe98e2277a51e0b3fb6e202b060aba04780fa07576"
},
"downloads": -1,
"filename": "yoyodyne-0.2.9.tar.gz",
"has_sig": false,
"md5_digest": "da869c009d74806121a207ab0a2fcce7",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 61352,
"upload_time": "2024-03-06T19:44:00",
"upload_time_iso_8601": "2024-03-06T19:44:00.199683Z",
"url": "https://files.pythonhosted.org/packages/65/0b/846e0e8ceaf43e5fc212f50bf9e5a97fbcd3c2b71525d0ba7cf4610b333c/yoyodyne-0.2.9.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-03-06 19:44:00",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "CUNY-CL",
"github_project": "yoyodyne",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"circle": true,
"requirements": [
{
"name": "black",
"specs": [
[
">=",
"23.3.0"
]
]
},
{
"name": "build",
"specs": [
[
">=",
"0.10.0"
]
]
},
{
"name": "flake8",
"specs": [
[
">=",
"6.0.0"
]
]
},
{
"name": "maxwell",
"specs": [
[
">=",
"0.2.0"
]
]
},
{
"name": "numpy",
"specs": [
[
">=",
"1.24.3"
]
]
},
{
"name": "pandas",
"specs": [
[
">=",
"1.5.3"
]
]
},
{
"name": "pytest",
"specs": [
[
">=",
"7.3.1"
]
]
},
{
"name": "pytorch-lightning",
"specs": [
[
"<",
"2.0.0"
],
[
">=",
"1.7.0"
]
]
},
{
"name": "scipy",
"specs": [
[
">=",
"1.10.1"
]
]
},
{
"name": "setuptools",
"specs": [
[
">=",
"67.8.0"
]
]
},
{
"name": "torch",
"specs": [
[
"<",
"2.0.0"
],
[
">=",
"1.11.0"
]
]
},
{
"name": "tqdm",
"specs": [
[
">=",
"4.65.0"
]
]
},
{
"name": "twine",
"specs": [
[
">=",
"4.0.2"
]
]
},
{
"name": "wandb",
"specs": [
[
">=",
"0.15.3"
]
]
},
{
"name": "wheel",
"specs": [
[
">=",
"0.40.0"
]
]
}
],
"lcname": "yoyodyne"
}
JSON
