facebookresearch--audiocraft
Alexandre Défossez (autodoc)
538 行
32 KiB
HTML
538 行
32 KiB
HTML
<!doctype html>
|
||
<html lang="en">
|
||
<head>
|
||
<meta charset="utf-8">
|
||
<meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
|
||
<meta name="generator" content="pdoc3 0.11.5">
|
||
<title>audiocraft.solvers.jasco API documentation</title>
|
||
<meta name="description" content="">
|
||
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
|
||
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/typography.min.css" integrity="sha512-Y1DYSb995BAfxobCkKepB1BqJJTPrOp3zPL74AWFugHHmmdcvO+C48WLrUOlhGMc0QG7AE3f7gmvvcrmX2fDoA==" crossorigin>
|
||
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/styles/default.min.css" crossorigin>
|
||
<style>:root{--highlight-color:#fe9}.flex{display:flex !important}body{line-height:1.5em}#content{padding:20px}#sidebar{padding:1.5em;overflow:hidden}#sidebar > *:last-child{margin-bottom:2cm}.http-server-breadcrumbs{font-size:130%;margin:0 0 15px 0}#footer{font-size:.75em;padding:5px 30px;border-top:1px solid #ddd;text-align:right}#footer p{margin:0 0 0 1em;display:inline-block}#footer p:last-child{margin-right:30px}h1,h2,h3,h4,h5{font-weight:300}h1{font-size:2.5em;line-height:1.1em}h2{font-size:1.75em;margin:2em 0 .50em 0}h3{font-size:1.4em;margin:1.6em 0 .7em 0}h4{margin:0;font-size:105%}h1:target,h2:target,h3:target,h4:target,h5:target,h6:target{background:var(--highlight-color);padding:.2em 0}a{color:#058;text-decoration:none;transition:color .2s ease-in-out}a:visited{color:#503}a:hover{color:#b62}.title code{font-weight:bold}h2[id^="header-"]{margin-top:2em}.ident{color:#900;font-weight:bold}pre code{font-size:.8em;line-height:1.4em;padding:1em;display:block}code{background:#f3f3f3;font-family:"DejaVu Sans Mono",monospace;padding:1px 4px;overflow-wrap:break-word}h1 code{background:transparent}pre{border-top:1px solid #ccc;border-bottom:1px solid #ccc;margin:1em 0}#http-server-module-list{display:flex;flex-flow:column}#http-server-module-list div{display:flex}#http-server-module-list dt{min-width:10%}#http-server-module-list p{margin-top:0}.toc ul,#index{list-style-type:none;margin:0;padding:0}#index code{background:transparent}#index h3{border-bottom:1px solid #ddd}#index ul{padding:0}#index h4{margin-top:.6em;font-weight:bold}@media (min-width:200ex){#index .two-column{column-count:2}}@media (min-width:300ex){#index .two-column{column-count:3}}dl{margin-bottom:2em}dl dl:last-child{margin-bottom:4em}dd{margin:0 0 1em 3em}#header-classes + dl > dd{margin-bottom:3em}dd dd{margin-left:2em}dd p{margin:10px 0}.name{background:#eee;font-size:.85em;padding:5px 10px;display:inline-block;min-width:40%}.name:hover{background:#e0e0e0}dt:target .name{background:var(--highlight-color)}.name > span:first-child{white-space:nowrap}.name.class > span:nth-child(2){margin-left:.4em}.inherited{color:#999;border-left:5px solid #eee;padding-left:1em}.inheritance em{font-style:normal;font-weight:bold}.desc h2{font-weight:400;font-size:1.25em}.desc h3{font-size:1em}.desc dt code{background:inherit}.source > summary,.git-link-div{color:#666;text-align:right;font-weight:400;font-size:.8em;text-transform:uppercase}.source summary > *{white-space:nowrap;cursor:pointer}.git-link{color:inherit;margin-left:1em}.source pre{max-height:500px;overflow:auto;margin:0}.source pre code{font-size:12px;overflow:visible;min-width:max-content}.hlist{list-style:none}.hlist li{display:inline}.hlist li:after{content:',\2002'}.hlist li:last-child:after{content:none}.hlist .hlist{display:inline;padding-left:1em}img{max-width:100%}td{padding:0 .5em}.admonition{padding:.1em 1em;margin:1em 0}.admonition-title{font-weight:bold}.admonition.note,.admonition.info,.admonition.important{background:#aef}.admonition.todo,.admonition.versionadded,.admonition.tip,.admonition.hint{background:#dfd}.admonition.warning,.admonition.versionchanged,.admonition.deprecated{background:#fd4}.admonition.error,.admonition.danger,.admonition.caution{background:lightpink}</style>
|
||
<style media="screen and (min-width: 700px)">@media screen and (min-width:700px){#sidebar{width:30%;height:100vh;overflow:auto;position:sticky;top:0}#content{width:70%;max-width:100ch;padding:3em 4em;border-left:1px solid #ddd}pre code{font-size:1em}.name{font-size:1em}main{display:flex;flex-direction:row-reverse;justify-content:flex-end}.toc ul ul,#index ul ul{padding-left:1em}.toc > ul > li{margin-top:.5em}}</style>
|
||
<style media="print">@media print{#sidebar h1{page-break-before:always}.source{display:none}}@media print{*{background:transparent !important;color:#000 !important;box-shadow:none !important;text-shadow:none !important}a[href]:after{content:" (" attr(href) ")";font-size:90%}a[href][title]:after{content:none}abbr[title]:after{content:" (" attr(title) ")"}.ir a:after,a[href^="javascript:"]:after,a[href^="#"]:after{content:""}pre,blockquote{border:1px solid #999;page-break-inside:avoid}thead{display:table-header-group}tr,img{page-break-inside:avoid}img{max-width:100% !important}@page{margin:0.5cm}p,h2,h3{orphans:3;widows:3}h1,h2,h3,h4,h5,h6{page-break-after:avoid}}</style>
|
||
<script defer src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/11.9.0/highlight.min.js" integrity="sha512-D9gUyxqja7hBtkWpPWGt9wfbfaMGVt9gnyCvYa+jojwwPHLCzUm5i8rpk7vD7wNee9bA35eYIjobYPaQuKS1MQ==" crossorigin></script>
|
||
<script>window.addEventListener('DOMContentLoaded', () => {
|
||
hljs.configure({languages: ['bash', 'css', 'diff', 'graphql', 'ini', 'javascript', 'json', 'plaintext', 'python', 'python-repl', 'rust', 'shell', 'sql', 'typescript', 'xml', 'yaml']});
|
||
hljs.highlightAll();
|
||
/* Collapse source docstrings */
|
||
setTimeout(() => {
|
||
[...document.querySelectorAll('.hljs.language-python > .hljs-string')]
|
||
.filter(el => el.innerHTML.length > 200 && ['"""', "'''"].includes(el.innerHTML.substring(0, 3)))
|
||
.forEach(el => {
|
||
let d = document.createElement('details');
|
||
d.classList.add('hljs-string');
|
||
d.innerHTML = '<summary>"""</summary>' + el.innerHTML.substring(3);
|
||
el.replaceWith(d);
|
||
});
|
||
}, 100);
|
||
})</script>
|
||
</head>
|
||
<body>
|
||
<main>
|
||
<article id="content">
|
||
<header>
|
||
<h1 class="title">Module <code>audiocraft.solvers.jasco</code></h1>
|
||
</header>
|
||
<section id="section-intro">
|
||
</section>
|
||
<section>
|
||
</section>
|
||
<section>
|
||
</section>
|
||
<section>
|
||
</section>
|
||
<section>
|
||
<h2 class="section-title" id="header-classes">Classes</h2>
|
||
<dl>
|
||
<dt id="audiocraft.solvers.jasco.JascoSolver"><code class="flex name class">
|
||
<span>class <span class="ident">JascoSolver</span></span>
|
||
<span>(</span><span>cfg: omegaconf.dictconfig.DictConfig)</span>
|
||
</code></dt>
|
||
<dd>
|
||
<details class="source">
|
||
<summary>
|
||
<span>Expand source code</span>
|
||
</summary>
|
||
<pre><code class="python">class JascoSolver(musicgen.MusicGenSolver):
|
||
"""Solver for JASCO - Joint Audio and Symbolic Conditioning for Temporally Controlled Text-to-Music Generation
|
||
https://arxiv.org/abs/2406.10970.
|
||
"""
|
||
DATASET_TYPE: builders.DatasetType = builders.DatasetType.JASCO
|
||
|
||
def __init__(self, cfg: DictConfig):
|
||
super().__init__(cfg)
|
||
|
||
# initialize generation parameters by config
|
||
self.generation_params = {
|
||
'cfg_coef_all': self.cfg.generate.lm.cfg_coef_all,
|
||
'cfg_coef_txt': self.cfg.generate.lm.cfg_coef_txt
|
||
}
|
||
|
||
self.latent_mean = cfg.compression_model_latent_mean
|
||
self.latent_std = cfg.compression_model_latent_std
|
||
self.mse = torch.nn.MSELoss(reduction='none')
|
||
self._best_metric_name = 'loss'
|
||
|
||
def build_model(self) -> None:
|
||
"""Instantiate model and optimization."""
|
||
assert self.cfg.efficient_attention_backend == "xformers", "JASCO v1 models support only xformers backend."
|
||
|
||
self.compression_model = CompressionSolver.wrapped_model_from_checkpoint(
|
||
self.cfg, self.cfg.compression_model_checkpoint, device=self.device)
|
||
assert self.compression_model.sample_rate == self.cfg.sample_rate, (
|
||
f"Compression model sample rate is {self.compression_model.sample_rate} but "
|
||
f"Solver sample rate is {self.cfg.sample_rate}."
|
||
)
|
||
# instantiate JASCO model
|
||
self.model: models.FlowMatchingModel = models.builders.get_jasco_model(self.cfg,
|
||
self.compression_model).to(self.device)
|
||
# initialize optimization
|
||
self.initialize_optimization()
|
||
|
||
def _get_latents(self, audio):
|
||
with torch.no_grad():
|
||
latents = self.compression_model.model.encoder(audio)
|
||
return latents.permute(0, 2, 1) # [B, D, T] -> [B, T, D]
|
||
|
||
def _prepare_latents_and_attributes(
|
||
self, batch: tp.Tuple[torch.Tensor, tp.List[SegmentWithAttributes]],
|
||
) -> tp.Tuple[dict, torch.Tensor, torch.Tensor]:
|
||
"""Prepare input batchs for language model training.
|
||
|
||
Args:
|
||
batch (tuple[torch.Tensor, list[SegmentWithAttributes]]): Input batch with audio tensor of shape [B, C, T]
|
||
and corresponding metadata as SegmentWithAttributes (with B items).
|
||
Returns:
|
||
Condition tensors (dict[str, any]): Preprocessed condition attributes.
|
||
Tokens (torch.Tensor): Audio tokens from compression model, of shape [B, K, T_s],
|
||
with B the batch size, K the number of codebooks, T_s the token timesteps.
|
||
Padding mask (torch.Tensor): Mask with valid positions in the tokens tensor, of shape [B, K, T_s].
|
||
"""
|
||
audio, infos = batch
|
||
audio = audio.to(self.device)
|
||
assert audio.size(0) == len(infos), (
|
||
f"Mismatch between number of items in audio batch ({audio.size(0)})",
|
||
f" and in metadata ({len(infos)})"
|
||
)
|
||
|
||
latents = self._get_latents(audio)
|
||
|
||
# prepare attributes
|
||
if JascoCondConst.CRD.value in self.cfg.conditioners:
|
||
null_chord_idx = self.cfg.conditioners.chords.chords_emb.card
|
||
else:
|
||
null_chord_idx = -1
|
||
attributes = [info.to_condition_attributes() for info in infos]
|
||
if self.model.cfg_dropout is not None:
|
||
attributes = self.model.cfg_dropout(samples=attributes,
|
||
cond_types=["wav", "text", "symbolic"],
|
||
null_chord_idx=null_chord_idx)
|
||
attributes = self.model.att_dropout(attributes)
|
||
tokenized = self.model.condition_provider.tokenize(attributes)
|
||
|
||
with self.autocast:
|
||
condition_tensors = self.model.condition_provider(tokenized)
|
||
|
||
# create a padding mask to hold valid vs invalid positions
|
||
padding_mask = torch.ones_like(latents, dtype=torch.bool, device=latents.device)
|
||
|
||
return condition_tensors, latents, padding_mask
|
||
|
||
def _normalized_latents(self, latents: torch.Tensor) -> torch.Tensor:
|
||
"""Normalize latents."""
|
||
return (latents - self.latent_mean) / self.latent_std
|
||
|
||
def _unnormalized_latents(self, latents: torch.Tensor) -> torch.Tensor:
|
||
"""Unnormalize latents."""
|
||
return (latents * self.latent_std) + self.latent_mean
|
||
|
||
def _z(self, z_0: torch.Tensor, z_1: torch.Tensor, t: torch.Tensor, sigma_min: float = 1e-5) -> torch.Tensor:
|
||
"""Interpolate data and prior."""
|
||
return (1 - (1 - sigma_min) * t) * z_0 + t * z_1
|
||
|
||
def _vector_field(self, z_0: torch.Tensor, z_1: torch.Tensor, sigma_min: float = 1e-5) -> torch.Tensor:
|
||
"""Compute the GT vector field.
|
||
sigma_min is a small value to avoid numerical instabilities."""
|
||
return z_1 - (1 - sigma_min) * z_0
|
||
|
||
def _compute_loss(self, t: torch.Tensor, v_theta: torch.Tensor, v: torch.Tensor) -> torch.Tensor:
|
||
"""Compute the loss."""
|
||
loss_func = self.cfg.get('loss_func', 'increasing')
|
||
if loss_func == 'uniform':
|
||
scales = 1
|
||
elif loss_func == 'increasing':
|
||
scales = 1 + t # type: ignore
|
||
elif loss_func == 'decreasing':
|
||
scales = 2 - t # type: ignore
|
||
else:
|
||
raise ValueError('unsupported loss_func was passed in config')
|
||
return (scales * self.mse(v_theta, v)).mean()
|
||
|
||
def run_step(self, idx: int, batch: tp.Tuple[torch.Tensor, tp.List[SegmentWithAttributes]], metrics: dict) -> dict:
|
||
"""Perform one training or valid step on a given batch."""
|
||
|
||
condition_tensors, latents, padding_mask = self._prepare_latents_and_attributes(batch)
|
||
|
||
self.deadlock_detect.update('tokens_and_conditions')
|
||
|
||
B, T, D = latents.shape
|
||
device = self.device
|
||
|
||
# normalize latents
|
||
z_1 = self._normalized_latents(latents)
|
||
|
||
# sample the N(0,1) prior
|
||
z_0 = torch.randn(B, T, D, device=device)
|
||
|
||
# random time parameter, between 0 to 1
|
||
t = torch.rand((B, 1, 1), device=device)
|
||
|
||
# interpolate data and prior
|
||
z = self._z(z_0, z_1, t)
|
||
|
||
# compute the GT vector field
|
||
v = self._vector_field(z_0, z_1)
|
||
|
||
with self.autocast:
|
||
v_theta = self.model(latents=z,
|
||
t=t,
|
||
conditions=[],
|
||
condition_tensors=condition_tensors)
|
||
|
||
loss = self._compute_loss(t, v_theta, v)
|
||
unscaled_loss = loss.clone()
|
||
|
||
self.deadlock_detect.update('loss')
|
||
|
||
if self.is_training:
|
||
metrics['lr'] = self.optimizer.param_groups[0]['lr']
|
||
if self.scaler is not None:
|
||
loss = self.scaler.scale(loss)
|
||
self.deadlock_detect.update('scale')
|
||
if self.cfg.fsdp.use:
|
||
loss.backward()
|
||
flashy.distrib.average_tensors(self.model.buffers())
|
||
elif self.cfg.optim.eager_sync:
|
||
with flashy.distrib.eager_sync_model(self.model):
|
||
loss.backward()
|
||
else:
|
||
# this should always be slower but can be useful
|
||
# for weird use cases like multiple backwards.
|
||
loss.backward()
|
||
flashy.distrib.sync_model(self.model)
|
||
self.deadlock_detect.update('backward')
|
||
|
||
if self.scaler is not None:
|
||
self.scaler.unscale_(self.optimizer)
|
||
if self.cfg.optim.max_norm:
|
||
if self.cfg.fsdp.use:
|
||
metrics['grad_norm'] = self.model.clip_grad_norm_(self.cfg.optim.max_norm) # type: ignore
|
||
else:
|
||
metrics['grad_norm'] = torch.nn.utils.clip_grad_norm_(
|
||
self.model.parameters(), self.cfg.optim.max_norm
|
||
)
|
||
if self.scaler is None:
|
||
self.optimizer.step()
|
||
else:
|
||
self.scaler.step(self.optimizer)
|
||
self.scaler.update()
|
||
if self.lr_scheduler:
|
||
self.lr_scheduler.step()
|
||
self.optimizer.zero_grad()
|
||
self.deadlock_detect.update('optim')
|
||
if self.scaler is not None:
|
||
scale = self.scaler.get_scale()
|
||
metrics['grad_scale'] = scale
|
||
if not loss.isfinite().all():
|
||
raise RuntimeError("Model probably diverged.")
|
||
|
||
metrics['loss'] = unscaled_loss
|
||
|
||
return metrics
|
||
|
||
def _decode_latents(self, latents):
|
||
return self.compression_model.model.decoder(latents.permute(0, 2, 1))
|
||
|
||
@torch.no_grad()
|
||
def run_generate_step(self, batch: tp.Tuple[torch.Tensor, tp.List[SegmentWithAttributes]],
|
||
gen_duration: float, prompt_duration: tp.Optional[float] = None,
|
||
remove_text_conditioning: bool = False,
|
||
**generation_params) -> dict:
|
||
"""Run generate step on a batch of optional audio tensor and corresponding attributes.
|
||
|
||
Args:
|
||
batch (tuple[torch.Tensor, list[SegmentWithAttributes]]):
|
||
use_prompt (bool): Whether to do audio continuation generation with prompt from audio batch.
|
||
gen_duration (float): Target audio duration for the generation.
|
||
prompt_duration (float, optional): Duration for the audio prompt to use for continuation.
|
||
remove_text_conditioning (bool, optional): Whether to remove the prompt from the generated audio.
|
||
generation_params: Additional generation parameters.
|
||
Returns:
|
||
gen_outputs (dict): Generation outputs, consisting in audio, audio tokens from both the generation
|
||
and the prompt along with additional information.
|
||
"""
|
||
bench_start = time.time()
|
||
audio, meta = batch
|
||
assert audio.size(0) == len(meta), (
|
||
f"Mismatch between number of items in audio batch ({audio.size(0)})",
|
||
f" and in metadata ({len(meta)})"
|
||
)
|
||
# prepare attributes
|
||
attributes = [x.to_condition_attributes() for x in meta]
|
||
|
||
# prepare audio prompt
|
||
if prompt_duration is None:
|
||
prompt_audio = None
|
||
else:
|
||
assert prompt_duration < gen_duration, "Prompt duration must be lower than target generation duration"
|
||
prompt_audio_frames = int(prompt_duration * self.compression_model.sample_rate)
|
||
prompt_audio = audio[..., :prompt_audio_frames]
|
||
|
||
# get audio tokens from compression model
|
||
if prompt_audio is None or prompt_audio.nelement() == 0:
|
||
num_samples = len(attributes)
|
||
prompt_tokens = None
|
||
else:
|
||
num_samples = None
|
||
prompt_audio = prompt_audio.to(self.device)
|
||
prompt_tokens, scale = self.compression_model.encode(prompt_audio)
|
||
assert scale is None, "Compression model in MusicGen should not require rescaling."
|
||
|
||
# generate by sampling from the LM
|
||
with self.autocast:
|
||
total_gen_len = math.ceil(gen_duration * self.compression_model.frame_rate)
|
||
gen_latents = self.model.generate(
|
||
prompt_tokens, attributes, max_gen_len=total_gen_len,
|
||
num_samples=num_samples, **self.generation_params)
|
||
|
||
# generate audio from latents
|
||
assert gen_latents.dim() == 3 # [B, T, D]
|
||
|
||
# unnormalize latents
|
||
gen_latents = self._unnormalized_latents(gen_latents)
|
||
gen_audio = self._decode_latents(gen_latents)
|
||
|
||
bench_end = time.time()
|
||
gen_outputs = {
|
||
'rtf': (bench_end - bench_start) / gen_duration,
|
||
'ref_audio': audio,
|
||
'gen_audio': gen_audio,
|
||
'gen_tokens': gen_latents,
|
||
'prompt_audio': prompt_audio,
|
||
'prompt_tokens': prompt_tokens,
|
||
}
|
||
return gen_outputs</code></pre>
|
||
</details>
|
||
<div class="desc"><p>Solver for JASCO - Joint Audio and Symbolic Conditioning for Temporally Controlled Text-to-Music Generation
|
||
<a href="https://arxiv.org/abs/2406.10970.">https://arxiv.org/abs/2406.10970.</a></p></div>
|
||
<h3>Ancestors</h3>
|
||
<ul class="hlist">
|
||
<li><a title="audiocraft.solvers.musicgen.MusicGenSolver" href="musicgen.html#audiocraft.solvers.musicgen.MusicGenSolver">MusicGenSolver</a></li>
|
||
<li><a title="audiocraft.solvers.base.StandardSolver" href="base.html#audiocraft.solvers.base.StandardSolver">StandardSolver</a></li>
|
||
<li>abc.ABC</li>
|
||
<li>flashy.solver.BaseSolver</li>
|
||
</ul>
|
||
<h3>Class variables</h3>
|
||
<dl>
|
||
<dt id="audiocraft.solvers.jasco.JascoSolver.DATASET_TYPE"><code class="name">var <span class="ident">DATASET_TYPE</span> : <a title="audiocraft.solvers.builders.DatasetType" href="builders.html#audiocraft.solvers.builders.DatasetType">DatasetType</a></code></dt>
|
||
<dd>
|
||
<div class="desc"></div>
|
||
</dd>
|
||
</dl>
|
||
<h3>Methods</h3>
|
||
<dl>
|
||
<dt id="audiocraft.solvers.jasco.JascoSolver.build_model"><code class="name flex">
|
||
<span>def <span class="ident">build_model</span></span>(<span>self) ‑> None</span>
|
||
</code></dt>
|
||
<dd>
|
||
<details class="source">
|
||
<summary>
|
||
<span>Expand source code</span>
|
||
</summary>
|
||
<pre><code class="python">def build_model(self) -> None:
|
||
"""Instantiate model and optimization."""
|
||
assert self.cfg.efficient_attention_backend == "xformers", "JASCO v1 models support only xformers backend."
|
||
|
||
self.compression_model = CompressionSolver.wrapped_model_from_checkpoint(
|
||
self.cfg, self.cfg.compression_model_checkpoint, device=self.device)
|
||
assert self.compression_model.sample_rate == self.cfg.sample_rate, (
|
||
f"Compression model sample rate is {self.compression_model.sample_rate} but "
|
||
f"Solver sample rate is {self.cfg.sample_rate}."
|
||
)
|
||
# instantiate JASCO model
|
||
self.model: models.FlowMatchingModel = models.builders.get_jasco_model(self.cfg,
|
||
self.compression_model).to(self.device)
|
||
# initialize optimization
|
||
self.initialize_optimization()</code></pre>
|
||
</details>
|
||
<div class="desc"><p>Instantiate model and optimization.</p></div>
|
||
</dd>
|
||
<dt id="audiocraft.solvers.jasco.JascoSolver.run_generate_step"><code class="name flex">
|
||
<span>def <span class="ident">run_generate_step</span></span>(<span>self,<br>batch: Tuple[torch.Tensor, List[<a title="audiocraft.modules.conditioners.SegmentWithAttributes" href="../modules/conditioners.html#audiocraft.modules.conditioners.SegmentWithAttributes">SegmentWithAttributes</a>]],<br>gen_duration: float,<br>prompt_duration: float | None = None,<br>remove_text_conditioning: bool = False,<br>**generation_params) ‑> dict</span>
|
||
</code></dt>
|
||
<dd>
|
||
<details class="source">
|
||
<summary>
|
||
<span>Expand source code</span>
|
||
</summary>
|
||
<pre><code class="python">@torch.no_grad()
|
||
def run_generate_step(self, batch: tp.Tuple[torch.Tensor, tp.List[SegmentWithAttributes]],
|
||
gen_duration: float, prompt_duration: tp.Optional[float] = None,
|
||
remove_text_conditioning: bool = False,
|
||
**generation_params) -> dict:
|
||
"""Run generate step on a batch of optional audio tensor and corresponding attributes.
|
||
|
||
Args:
|
||
batch (tuple[torch.Tensor, list[SegmentWithAttributes]]):
|
||
use_prompt (bool): Whether to do audio continuation generation with prompt from audio batch.
|
||
gen_duration (float): Target audio duration for the generation.
|
||
prompt_duration (float, optional): Duration for the audio prompt to use for continuation.
|
||
remove_text_conditioning (bool, optional): Whether to remove the prompt from the generated audio.
|
||
generation_params: Additional generation parameters.
|
||
Returns:
|
||
gen_outputs (dict): Generation outputs, consisting in audio, audio tokens from both the generation
|
||
and the prompt along with additional information.
|
||
"""
|
||
bench_start = time.time()
|
||
audio, meta = batch
|
||
assert audio.size(0) == len(meta), (
|
||
f"Mismatch between number of items in audio batch ({audio.size(0)})",
|
||
f" and in metadata ({len(meta)})"
|
||
)
|
||
# prepare attributes
|
||
attributes = [x.to_condition_attributes() for x in meta]
|
||
|
||
# prepare audio prompt
|
||
if prompt_duration is None:
|
||
prompt_audio = None
|
||
else:
|
||
assert prompt_duration < gen_duration, "Prompt duration must be lower than target generation duration"
|
||
prompt_audio_frames = int(prompt_duration * self.compression_model.sample_rate)
|
||
prompt_audio = audio[..., :prompt_audio_frames]
|
||
|
||
# get audio tokens from compression model
|
||
if prompt_audio is None or prompt_audio.nelement() == 0:
|
||
num_samples = len(attributes)
|
||
prompt_tokens = None
|
||
else:
|
||
num_samples = None
|
||
prompt_audio = prompt_audio.to(self.device)
|
||
prompt_tokens, scale = self.compression_model.encode(prompt_audio)
|
||
assert scale is None, "Compression model in MusicGen should not require rescaling."
|
||
|
||
# generate by sampling from the LM
|
||
with self.autocast:
|
||
total_gen_len = math.ceil(gen_duration * self.compression_model.frame_rate)
|
||
gen_latents = self.model.generate(
|
||
prompt_tokens, attributes, max_gen_len=total_gen_len,
|
||
num_samples=num_samples, **self.generation_params)
|
||
|
||
# generate audio from latents
|
||
assert gen_latents.dim() == 3 # [B, T, D]
|
||
|
||
# unnormalize latents
|
||
gen_latents = self._unnormalized_latents(gen_latents)
|
||
gen_audio = self._decode_latents(gen_latents)
|
||
|
||
bench_end = time.time()
|
||
gen_outputs = {
|
||
'rtf': (bench_end - bench_start) / gen_duration,
|
||
'ref_audio': audio,
|
||
'gen_audio': gen_audio,
|
||
'gen_tokens': gen_latents,
|
||
'prompt_audio': prompt_audio,
|
||
'prompt_tokens': prompt_tokens,
|
||
}
|
||
return gen_outputs</code></pre>
|
||
</details>
|
||
<div class="desc"><p>Run generate step on a batch of optional audio tensor and corresponding attributes.</p>
|
||
<h2 id="args">Args</h2>
|
||
<dl>
|
||
<dt>batch (tuple[torch.Tensor, list[SegmentWithAttributes]]):</dt>
|
||
<dt><strong><code>use_prompt</code></strong> : <code>bool</code></dt>
|
||
<dd>Whether to do audio continuation generation with prompt from audio batch.</dd>
|
||
<dt><strong><code>gen_duration</code></strong> : <code>float</code></dt>
|
||
<dd>Target audio duration for the generation.</dd>
|
||
<dt><strong><code>prompt_duration</code></strong> : <code>float</code>, optional</dt>
|
||
<dd>Duration for the audio prompt to use for continuation.</dd>
|
||
<dt><strong><code>remove_text_conditioning</code></strong> : <code>bool</code>, optional</dt>
|
||
<dd>Whether to remove the prompt from the generated audio.</dd>
|
||
<dt><strong><code>generation_params</code></strong></dt>
|
||
<dd>Additional generation parameters.</dd>
|
||
</dl>
|
||
<h2 id="returns">Returns</h2>
|
||
<p>gen_outputs (dict): Generation outputs, consisting in audio, audio tokens from both the generation
|
||
and the prompt along with additional information.</p></div>
|
||
</dd>
|
||
</dl>
|
||
<h3>Inherited members</h3>
|
||
<ul class="hlist">
|
||
<li><code><b><a title="audiocraft.solvers.musicgen.MusicGenSolver" href="musicgen.html#audiocraft.solvers.musicgen.MusicGenSolver">MusicGenSolver</a></b></code>:
|
||
<ul class="hlist">
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.autocast" href="base.html#audiocraft.solvers.base.StandardSolver.autocast">autocast</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.best_metric_name" href="base.html#audiocraft.solvers.base.StandardSolver.best_metric_name">best_metric_name</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.build_dataloaders" href="musicgen.html#audiocraft.solvers.musicgen.MusicGenSolver.build_dataloaders">build_dataloaders</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.commit" href="base.html#audiocraft.solvers.base.StandardSolver.commit">commit</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.common_train_valid" href="base.html#audiocraft.solvers.base.StandardSolver.common_train_valid">common_train_valid</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.evaluate" href="base.html#audiocraft.solvers.base.StandardSolver.evaluate">evaluate</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.evaluate_audio_generation" href="musicgen.html#audiocraft.solvers.musicgen.MusicGenSolver.evaluate_audio_generation">evaluate_audio_generation</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.generate" href="base.html#audiocraft.solvers.base.StandardSolver.generate">generate</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.generate_audio" href="musicgen.html#audiocraft.solvers.musicgen.MusicGenSolver.generate_audio">generate_audio</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.get_eval_solver_from_sig" href="musicgen.html#audiocraft.solvers.musicgen.MusicGenSolver.get_eval_solver_from_sig">get_eval_solver_from_sig</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.initialize_ema" href="base.html#audiocraft.solvers.base.StandardSolver.initialize_ema">initialize_ema</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.load_checkpoints" href="base.html#audiocraft.solvers.base.StandardSolver.load_checkpoints">load_checkpoints</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.log_model_summary" href="base.html#audiocraft.solvers.base.StandardSolver.log_model_summary">log_model_summary</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.register_best_state" href="base.html#audiocraft.solvers.base.StandardSolver.register_best_state">register_best_state</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.register_ema" href="base.html#audiocraft.solvers.base.StandardSolver.register_ema">register_ema</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.restore" href="base.html#audiocraft.solvers.base.StandardSolver.restore">restore</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.run" href="base.html#audiocraft.solvers.base.StandardSolver.run">run</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.run_epoch" href="base.html#audiocraft.solvers.base.StandardSolver.run_epoch">run_epoch</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.run_one_stage" href="base.html#audiocraft.solvers.base.StandardSolver.run_one_stage">run_one_stage</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.run_step" href="base.html#audiocraft.solvers.base.StandardSolver.run_step">run_step</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.save_checkpoints" href="base.html#audiocraft.solvers.base.StandardSolver.save_checkpoints">save_checkpoints</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.should_run_stage" href="base.html#audiocraft.solvers.base.StandardSolver.should_run_stage">should_run_stage</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.should_stop_training" href="base.html#audiocraft.solvers.base.StandardSolver.should_stop_training">should_stop_training</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.show" href="musicgen.html#audiocraft.solvers.musicgen.MusicGenSolver.show">show</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.train" href="base.html#audiocraft.solvers.base.StandardSolver.train">train</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.update_best_state_from_stage" href="base.html#audiocraft.solvers.base.StandardSolver.update_best_state_from_stage">update_best_state_from_stage</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.musicgen.MusicGenSolver.valid" href="base.html#audiocraft.solvers.base.StandardSolver.valid">valid</a></code></li>
|
||
</ul>
|
||
</li>
|
||
</ul>
|
||
</dd>
|
||
</dl>
|
||
</section>
|
||
</article>
|
||
<nav id="sidebar">
|
||
<div class="toc">
|
||
<ul></ul>
|
||
</div>
|
||
<ul id="index">
|
||
<li><h3>Super-module</h3>
|
||
<ul>
|
||
<li><code><a title="audiocraft.solvers" href="index.html">audiocraft.solvers</a></code></li>
|
||
</ul>
|
||
</li>
|
||
<li><h3><a href="#header-classes">Classes</a></h3>
|
||
<ul>
|
||
<li>
|
||
<h4><code><a title="audiocraft.solvers.jasco.JascoSolver" href="#audiocraft.solvers.jasco.JascoSolver">JascoSolver</a></code></h4>
|
||
<ul class="">
|
||
<li><code><a title="audiocraft.solvers.jasco.JascoSolver.DATASET_TYPE" href="#audiocraft.solvers.jasco.JascoSolver.DATASET_TYPE">DATASET_TYPE</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.jasco.JascoSolver.build_model" href="#audiocraft.solvers.jasco.JascoSolver.build_model">build_model</a></code></li>
|
||
<li><code><a title="audiocraft.solvers.jasco.JascoSolver.run_generate_step" href="#audiocraft.solvers.jasco.JascoSolver.run_generate_step">run_generate_step</a></code></li>
|
||
</ul>
|
||
</li>
|
||
</ul>
|
||
</li>
|
||
</ul>
|
||
</nav>
|
||
</main>
|
||
<footer id="footer">
|
||
<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
|
||
</footer>
|
||
</body>
|
||
</html>
|