"""Defines interfaces for simple inference with pretrained models
Authors:
* Aku Rouhe 2021
* Peter Plantinga 2021
* Loren Lugosch 2020
* Mirco Ravanelli 2020
* Titouan Parcollet 2021
* Abdel Heba 2021
* Andreas Nautsch 2022, 2023
* Pooneh Mousavi 2023
* Sylvain de Langen 2023
* Adel Moumen 2023
* Pradnya Kandarkar 2023
"""
import sys
import warnings
from types import SimpleNamespace
import torch
from hyperpyyaml import load_hyperpyyaml
from torch.nn import (
DataParallel as DP,
SyncBatchNorm,
)
from torch.nn.parallel import DistributedDataParallel as DDP
from speechbrain.dataio import audio_io
from speechbrain.dataio.batch import PaddedBatch, PaddedData
from speechbrain.dataio.preprocess import AudioNormalizer
from speechbrain.utils.autocast import AMPConfig, TorchAutocast
from speechbrain.utils.data_pipeline import DataPipeline
from speechbrain.utils.data_utils import split_path
from speechbrain.utils.distributed import infer_device
from speechbrain.utils.fetching import FetchConfig, LocalStrategy, fetch
from speechbrain.utils.logger import get_logger
from speechbrain.utils.run_opts import RunOptions
from speechbrain.utils.superpowers import import_from_path
logger = get_logger(__name__)
[docs]
def foreign_class(
source,
hparams_file="hyperparams.yaml",
pymodule_file="custom.py",
classname="CustomInterface",
savedir=None,
local_strategy: LocalStrategy = LocalStrategy.SYMLINK,
fetch_config: FetchConfig = FetchConfig(),
**kwargs,
):
"""Thin wrapper for `pretrained_from_hparams()` that fetches and loads a custom class.
The pymodule file should contain a class with the given classname. An
instance of that class is returned. The idea is to have a custom Pretrained
subclass in the file. The pymodule file is also added to the python path
before the Hyperparams YAML file is loaded, so it can contain any custom
implementations that are needed.
.. warning::
Caution should be used with this function as it can download and run
arbitrary code onto the machine this function is used on. Only use
this function when the target module is from a highly trusted source!
Arguments
---------
source : str or Path or FetchSource
The location to use for finding the model. See
``speechbrain.utils.fetching.fetch`` for details.
hparams_file : str
The name of the hyperparameters file to use for constructing
the modules necessary for inference. Must contain two keys:
"modules" and "pretrainer", as described in `pretrained_from_hparams`.
pymodule_file : str
The name of the Python file containing the model's python class. The file
will be fetched from `source` and will be used to load the class code.
classname : str
The name of the model's Python class, which should be present in the
code of the `pymodule_file`.
savedir : Optional[Union[str, Path]]
Where to put the pretraining material. If not given, just use cache.
local_strategy : LocalStrategy, default LocalStrategy.SYMLINK
Type of caching to use for keeping a local copy.
fetch_config : FetchConfig
Configuration options for caching and other fetch behavior.
**kwargs
Arguments to pass to `pretrained_from_hparams`
Returns
-------
object
An instance of a class with the given classname from the given pymodule file.
"""
pymodule_local_path = fetch(
filename=pymodule_file,
source=source,
savedir=savedir,
save_filename=None,
local_strategy=local_strategy,
fetch_config=fetch_config,
)
sys.path.append(str(pymodule_local_path.parent))
# Dynamically import the specified Python module and retrieve the class by name.
# This allows users to define custom model interfaces outside of SpeechBrain.
# After importing, passes the class (not an instance) to pretrained_from_hparams,
# which will handle loading and instantiation with the appropriate hyperparameters.
module = import_from_path(pymodule_local_path)
cls = getattr(module, classname)
return pretrained_from_hparams(
cls=cls,
source=source,
hparams_file=hparams_file,
savedir=savedir,
local_strategy=local_strategy,
fetch_config=fetch_config,
**kwargs,
)
[docs]
def pretrained_from_hparams(
cls,
source,
hparams_file="hyperparams.yaml",
overrides={},
overrides_must_match=True,
savedir=None,
download_only=False,
local_strategy: LocalStrategy = LocalStrategy.SYMLINK,
fetch_config: FetchConfig = FetchConfig(),
**kwargs,
):
"""Fetch and load an interface from an outside source
The source can be a location on the filesystem or online/huggingface
The hyperparams file should contain a "modules" key, which is a
dictionary of torch modules used for computation.
The hyperparams file should contain a "pretrainer" key, which is a
speechbrain.utils.parameter_transfer.Pretrainer
.. warning::
Caution should be used with this function as it can download and run
arbitrary code onto the machine this function is used on. Only use
this function when the target hparams file is from a highly trusted source!
Arguments
---------
cls : Type[Pretrained]
The class to construct an instance of, usually a sub-type of Pretrained
source : str or Path or FetchSource
The location to use for finding the model. See
``speechbrain.utils.fetching.fetch`` for details.
hparams_file : str
The name of the hyperparameters file to use for constructing
the modules necessary for inference. Must contain two keys:
"modules" and "pretrainer", as described.
overrides : dict
Any changes to make to the hparams file when it is loaded.
overrides_must_match : bool
Whether an error will be thrown when an override does not match
a corresponding key in the yaml_stream.
savedir : str or Path
Where to put the pretraining material. If not given, just use cache.
download_only : bool (default: False)
If true, class and instance creation is skipped.
local_strategy : LocalStrategy, default LocalStrategy.SYMLINK
Type of caching to use for keeping a local copy.
fetch_config : FetchConfig
Configuration options for caching and other fetch behavior.
**kwargs : dict
Arguments to forward to class constructor.
Returns
-------
object : Optional[Pretrained]
An instance of a Pretrained class, constructed from the hparams.
None is returned if the argument `download_only` is `True`.
"""
hparams_local_path = fetch(
filename=hparams_file,
source=source,
savedir=savedir,
save_filename=None,
local_strategy=local_strategy,
fetch_config=fetch_config,
)
# Load the modules:
with open(hparams_local_path, encoding="utf-8") as fin:
hparams = load_hyperpyyaml(fin, overrides, overrides_must_match)
hparams["savedir"] = savedir
# Pretraining:
pretrainer = hparams["pretrainer"]
pretrainer.set_collect_in(savedir)
pretrainer.collect_files(
default_source=source,
local_strategy=local_strategy,
fetch_config=fetch_config,
)
# Load on the CPU. Later the params can be moved elsewhere by specifying
if not download_only:
# run_opts={"device": ...}
pretrainer.load_collected()
return cls(modules=hparams["modules"], hparams=hparams, **kwargs)
# Not strictly necessary, but let's be explicit here
else:
return None
[docs]
class Pretrained(torch.nn.Module):
"""Takes a trained model and makes predictions on new data.
This is a base class which handles some common boilerplate.
It intentionally has an interface similar to ``Brain`` - these base
classes handle similar things.
Subclasses of Pretrained should implement the actual logic of how
the pretrained system runs, and add methods with descriptive names
(e.g. transcribe_file() for ASR).
Pretrained is a torch.nn.Module so that methods like .to() or .eval() can
work. Subclasses should provide a suitable forward() implementation: by
convention, it should be a method that takes a batch of audio signals and
runs the full model (as applicable).
Arguments
---------
modules : dict of str:torch.nn.Module pairs
The Torch modules that make up the learned system. These can be treated
in special ways (put on the right device, frozen, etc.). These are available
as attributes under ``self.mods``, like self.mods.model(x)
hparams : dict
Each key:value pair should consist of a string key and a hyperparameter
that is used within the overridden methods. These will
be accessible via an ``hparams`` attribute, using "dot" notation:
e.g., self.hparams.model(x).
run_opts : Optional[Union[RunOptions, dict]]
A set of options to change the runtime environment, see ``RunOptions`` for
a complete list. Some options are meant for training, and will not apply
for this instance intended for inference.
freeze_params : bool
To freeze (requires_grad=False) parameters or not. Normally in inference
you want to freeze the params. Also calls .eval() on all modules.
"""
HPARAMS_NEEDED = []
MODULES_NEEDED = []
def __init__(
self, modules=None, hparams=None, run_opts=None, freeze_params=True
):
super().__init__()
# Check which options have been overridden. Order of priority
# is lowest: default < hparams < run_opts: highest
if isinstance(run_opts, dict):
run_opts = RunOptions.from_dictionary(run_opts)
self.run_opt_defaults = RunOptions()
for arg, default in self.run_opt_defaults.as_dict().items():
if run_opts is not None and arg in run_opts.overridden_args:
setattr(self, arg, run_opts[arg])
# If any arg from run_opt_defaults exist in hparams and
# not in command line args "run_opts"
elif hparams is not None and arg in hparams:
setattr(self, arg, hparams[arg])
else:
setattr(self, arg, default)
# If device was not provided, make a best guess
if self.device is None:
self.device = infer_device()
# Set device type based on device string
if self.device == "cpu":
self.device_type = "cpu"
elif "cuda" in self.device:
self.device_type = "cuda"
# Set cuda device based on device string
try:
_, device_index = self.device.split(":")
torch.cuda.set_device(int(device_index))
except (ValueError, IndexError, TypeError) as e:
logger.warning(
f"Could not parse CUDA device string '{self.device}': {e}. Falling back to device 0."
)
torch.cuda.set_device(0)
else:
self.device_type = self.device.split(":")[0]
precision_dtype = AMPConfig.from_name(self.precision).dtype
self.inference_ctx = TorchAutocast(
device_type=self.device_type, dtype=precision_dtype
)
# Put modules on the right device, accessible with dot notation
self.mods = torch.nn.ModuleDict(modules)
for module in self.mods.values():
if module is not None:
module.to(self.device)
# Check MODULES_NEEDED and HPARAMS_NEEDED and
# make hyperparams available with dot notation
if self.HPARAMS_NEEDED and hparams is None:
raise ValueError("Need to provide hparams dict.")
if hparams is not None:
# Also first check that all required params are found:
for hp in self.HPARAMS_NEEDED:
if hp not in hparams:
raise ValueError(f"Need hparams['{hp}']")
self.hparams = SimpleNamespace(**hparams)
# Prepare modules for computation, e.g. jit
self._prepare_modules(freeze_params)
# Audio normalization
self.audio_normalizer = hparams.get(
"audio_normalizer", AudioNormalizer()
)
def _prepare_modules(self, freeze_params):
"""Prepare modules for computation, e.g. jit.
Arguments
---------
freeze_params : bool
Whether to freeze the parameters and call ``eval()``.
"""
# Make jit-able
self._compile()
self._wrap_distributed()
# If we don't want to backprop, freeze the pretrained parameters
if freeze_params:
self.mods.eval()
for p in self.mods.parameters():
p.requires_grad = False
[docs]
def load_audio(self, path, savedir=None):
"""Load an audio file with this model's input spec
When using a speech model, it is important to use the same type of data,
as was used to train the model. This means for example using the same
sampling rate and number of channels. It is, however, possible to
convert a file from a higher sampling rate to a lower one (downsampling).
Similarly, it is simple to downmix a stereo file to mono.
The path can be a local path, a web url, or a link to a huggingface repo.
"""
source, fl = split_path(path)
path = fetch(fl, source=source, savedir=savedir)
signal, sr = audio_io.load(str(path), channels_first=False)
signal = signal.to(self.device)
return self.audio_normalizer(signal, sr)
def _compile(self):
"""Compile requested modules with either JIT or TorchInductor."""
compile_available = hasattr(torch, "compile")
if not compile_available and self.compile_module_keys is not None:
raise ValueError(
"'compile_module_keys' specified, but this install of PyTorch "
"seems to be too old to support it."
)
# Modules to compile with torch.compile
compile_module_keys = set()
if self.compile:
if self.compile_module_keys is None:
compile_module_keys = set(self.mods)
else:
compile_module_keys = set(self.compile_module_keys)
logger.warning(
"--compile and --compile_module_keys are both specified. "
"Only modules specified in --compile_module_keys will be compiled."
)
# Modules to compile with jit
jit_module_keys = set()
if self.jit:
if self.jit_module_keys is None:
jit_module_keys = set(self.mods)
else:
jit_module_keys = set(self.jit_module_keys)
logger.warning(
"--jit and --jit_module_keys are both specified. "
"Only modules specified in --jit_module_keys will be compiled."
)
# find missing keys
for name in compile_module_keys | jit_module_keys:
if name not in self.mods:
raise ValueError(
f"module {name} is not defined in your hparams file."
)
# try 'torch.compile', remove successful compiles from JIT list
for name in compile_module_keys:
try:
module = torch.compile(
self.mods[name],
mode=self.compile_mode,
fullgraph=self.compile_using_fullgraph,
dynamic=self.compile_using_dynamic_shape_tracing,
)
except Exception as e:
logger.warning(
f"'{name}' in 'compile_module_keys' failed to compile "
f"and will be skipped (may fallback onto JIT, if "
f"specified): {e}"
)
continue
self.mods[name] = module.to(self.device)
jit_module_keys.discard(name)
for name in jit_module_keys:
module = torch.jit.script(self.mods[name])
self.mods[name] = module.to(self.device)
def _compile_jit(self):
warnings.warn("'_compile_jit' is deprecated; use '_compile' instead")
self._compile()
def _wrap_distributed(self):
"""Wrap modules with distributed wrapper when requested."""
if not self.distributed_launch and not self.data_parallel_backend:
return
elif self.distributed_launch:
for name, module in self.mods.items():
if any(p.requires_grad for p in module.parameters()):
# for ddp, all module must run on same GPU
module = SyncBatchNorm.convert_sync_batchnorm(module)
module = DDP(module, device_ids=[self.device])
self.mods[name] = module
else:
# data_parallel_backend
for name, module in self.mods.items():
if any(p.requires_grad for p in module.parameters()):
# if distributed_count = -1 then use all gpus
# otherwise, specify the set of gpu to use
if self.data_parallel_count == -1:
module = DP(module)
else:
module = DP(
module, [i for i in range(self.data_parallel_count)]
)
self.mods[name] = module
[docs]
@classmethod
def from_hparams(cls, source, hparams_file="hyperparams.yaml", **kwargs):
"""Fetch and load based from outside source based on HyperPyYAML file
The source can be a location on the filesystem or online/huggingface
The hyperparams file should contain a "modules" key, which is a
dictionary of torch modules used for computation.
The hyperparams file should contain a "pretrainer" key, which is a
speechbrain.utils.parameter_transfer.Pretrainer
.. warning::
Caution should be used with this function as it can download and run
arbitrary code onto the machine this function is used on. Only use
this function when the target hparams file is from a highly trusted source!
Arguments
---------
source : str
The location to use for finding the model. See
``speechbrain.utils.fetching.fetch`` for details.
hparams_file : str
The name of the hyperparameters file to use for constructing
the modules necessary for inference. Must contain two keys:
"modules" and "pretrainer", as described.
**kwargs : dict
Arguments to forward to `pretrained_from_hparams`.
Returns
-------
Instance of cls
"""
return pretrained_from_hparams(
cls=cls, source=source, hparams_file=hparams_file, **kwargs
)
[docs]
class EncodeDecodePipelineMixin:
"""
A mixin for pretrained models that makes it possible to specify an encoding pipeline and a decoding pipeline
"""
[docs]
def create_pipelines(self):
"""
Initializes the encode and decode pipeline
"""
self._run_init_steps(self.hparams.encode_pipeline)
self._run_init_steps(self.hparams.decode_pipeline)
self.encode_pipeline = DataPipeline(
static_data_keys=self.INPUT_STATIC_KEYS,
dynamic_items=self.hparams.encode_pipeline["steps"],
output_keys=self.hparams.encode_pipeline["output_keys"],
)
self.decode_pipeline = DataPipeline(
static_data_keys=self.hparams.model_output_keys,
dynamic_items=self.hparams.decode_pipeline["steps"],
output_keys=self.OUTPUT_KEYS,
)
def _run_init_steps(self, pipeline_definition):
"""Encode/decode pipelines may include initialization
steps, such as filling text encoders with tokens. Calling
this method will run them, if defined"""
steps = pipeline_definition.get("init", [])
for step in steps:
step_func = step.get("func")
if not step_func or not callable(step_func):
raise ValueError("Invalid pipeline init definition")
step_func()
def _run_pipeline(self, pipeline, input, batch):
if batch:
output = pipeline(input)
else:
output = [pipeline(item) for item in input]
return output
def _get_encode_pipeline_input(self, input):
return input if self.batch_inputs else self._itemize(input)
def _get_decode_pipeline_input(self, model_output):
model_output_keys = getattr(self.hparams, "model_output_keys", None)
pipeline_input = model_output
if len(model_output_keys) == 1:
pipeline_input = (pipeline_input,)
# The input to a pipeline is a dictionary. If model_output_keys
# is provided, the output of the model is assumed to be a collection
# (e.g. a list or a tuple).
if model_output_keys:
pipeline_input = dict(zip(model_output_keys, pipeline_input))
# By default, the pipeline will be applied to in batch mode
# to the entire model input
if not self.batch_outputs:
pipeline_input = self._itemize(pipeline_input)
return pipeline_input
def _itemize(self, pipeline_input):
first_item = next(iter(pipeline_input.values()))
keys, values = pipeline_input.keys(), pipeline_input.values()
batch_length = len(first_item)
return [
dict(zip(keys, [value[idx] for value in values]))
for idx in range(batch_length)
]
[docs]
def to_dict(self, data):
"""
Converts padded batches to dictionaries, leaves
other data types as is
Arguments
---------
data: object
a dictionary or a padded batch
Returns
-------
results: dict
the dictionary
"""
if isinstance(data, PaddedBatch):
data = {
key: self._get_value(data, key)
for key in self.hparams.encode_pipeline["output_keys"]
}
return data
def _get_value(self, data, key):
"""
Retrieves the value associated with the specified key, dereferencing
.data where applicable
Arguments
---------
data: PaddedBatch
a padded batch
key: str
the key
Returns
-------
result: object
the result
"""
value = getattr(data, key)
if not self.input_use_padded_data and isinstance(value, PaddedData):
value = value.data
return value
@property
def batch_inputs(self):
"""
Determines whether the input pipeline
operates on batches or individual examples
(true means batched)
Returns
-------
batch_inputs: bool
"""
return self.hparams.encode_pipeline.get("batch", True)
@property
def input_use_padded_data(self):
"""
If turned on, raw PaddedData instances will be passed to
the model. If turned off, only .data will be used
Returns
-------
result: bool
whether padded data is used as is
"""
return self.hparams.encode_pipeline.get("use_padded_data", False)
@property
def batch_outputs(self):
"""
Determines whether the output pipeline
operates on batches or individual examples
(true means batched)
Returns
-------
batch_outputs: bool
"""
return self.hparams.decode_pipeline.get("batch", True)
def _collate(self, data):
if not self.batch_inputs:
collate_fn = getattr(self.hparams, "collate_fn", PaddedBatch)
data = collate_fn(data)
return data
[docs]
def decode_output(self, output):
"""
Decodes the raw model outputs
Arguments
---------
output: tuple
raw model outputs
Returns
-------
result: dict or list
the output of the pipeline
"""
pipeline_input = self._get_decode_pipeline_input(output)
return self._run_pipeline(
pipeline=self.decode_pipeline,
input=pipeline_input,
batch=self.batch_outputs,
)