Upload 5 files

audiocraft/grids/compression/_explorers.py

@@ -0,0 +1,55 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+import treetable as tt
+
+from .._base_explorers import BaseExplorer
+
+
+class CompressionExplorer(BaseExplorer):
+    eval_metrics = ["sisnr", "visqol"]
+
+    def stages(self):
+        return ["train", "valid", "evaluate"]
+
+    def get_grid_meta(self):
+        """Returns the list of Meta information to display for each XP/job.
+        """
+        return [
+            tt.leaf("index", align=">"),
+            tt.leaf("name", wrap=140),
+            tt.leaf("state"),
+            tt.leaf("sig", align=">"),
+        ]
+
+    def get_grid_metrics(self):
+        """Return the metrics that should be displayed in the tracking table.
+        """
+        return [
+            tt.group(
+                "train",
+                [
+                    tt.leaf("epoch"),
+                    tt.leaf("bandwidth", ".2f"),
+                    tt.leaf("adv", ".4f"),
+                    tt.leaf("d_loss", ".4f"),
+                ],
+                align=">",
+            ),
+            tt.group(
+                "valid",
+                [
+                    tt.leaf("bandwidth", ".2f"),
+                    tt.leaf("adv", ".4f"),
+                    tt.leaf("msspec", ".4f"),
+                    tt.leaf("sisnr", ".2f"),
+                ],
+                align=">",
+            ),
+            tt.group(
+                "evaluate", [tt.leaf(name, ".3f") for name in self.eval_metrics], align=">"
+            ),
+        ]

audiocraft/grids/compression/debug.py

@@ -0,0 +1,31 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+"""
+Grid search file, simply list all the exp you want in `explorer`.
+Any new exp added there will be scheduled.
+You can cancel and experiment by commenting its line.
+
+This grid is a minimal example for debugging compression task
+and how to override parameters directly in a grid.
+Learn more about dora grids: https://github.com/facebookresearch/dora
+"""
+
+from ._explorers import CompressionExplorer
+from ...environment import AudioCraftEnvironment
+
+
+@CompressionExplorer
+def explorer(launcher):
+    partitions = AudioCraftEnvironment.get_slurm_partitions(['team', 'global'])
+    launcher.slurm_(gpus=2, partition=partitions)
+    launcher.bind_(solver='compression/debug')
+
+    with launcher.job_array():
+        # base debug task using config from solver=compression/debug
+        launcher()
+        # we can override parameters in the grid to launch additional xps
+        launcher({'rvq.bins': 2048, 'rvq.n_q': 4})

audiocraft/grids/compression/encodec_audiogen_16khz.py

@@ -0,0 +1,29 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+"""
+Grid search file, simply list all the exp you want in `explorer`.
+Any new exp added there will be scheduled.
+You can cancel and experiment by commenting its line.
+
+This grid shows how to train the new AudioGen EnCodec model at 16 kHz.
+"""
+
+from ._explorers import CompressionExplorer
+from ...environment import AudioCraftEnvironment
+
+
+@CompressionExplorer
+def explorer(launcher):
+    partitions = AudioCraftEnvironment.get_slurm_partitions(['team', 'global'])
+    launcher.slurm_(gpus=8, partition=partitions)
+    # use configuration for AudioGen's EnCodec model trained on monophonic audio sampled at 16 kHz
+    # AudioGen's EnCodec is trained with a total stride of 320 leading to a frame rate of 50 hz
+    launcher.bind_(solver='compression/encodec_audiogen_16khz')
+    # replace this by the desired sound dataset
+    launcher.bind_(dset='internal/sounds_16khz')
+    # launch xp
+    launcher()

audiocraft/grids/compression/encodec_base_24khz.py

@@ -0,0 +1,28 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+"""
+Grid search file, simply list all the exp you want in `explorer`.
+Any new exp added there will be scheduled.
+You can cancel and experiment by commenting its line.
+
+This grid shows how to train a base causal EnCodec model at 24 kHz.
+"""
+
+from ._explorers import CompressionExplorer
+from ...environment import AudioCraftEnvironment
+
+
+@CompressionExplorer
+def explorer(launcher):
+    partitions = AudioCraftEnvironment.get_slurm_partitions(['team', 'global'])
+    launcher.slurm_(gpus=8, partition=partitions)
+    # base causal EnCodec trained on monophonic audio sampled at 24 kHz
+    launcher.bind_(solver='compression/encodec_base_24khz')
+    # replace this by the desired dataset
+    launcher.bind_(dset='audio/example')
+    # launch xp
+    launcher()

audiocraft/grids/compression/encodec_musicgen_32khz.py

@@ -0,0 +1,34 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+"""
+Grid search file, simply list all the exp you want in `explorer`.
+Any new exp added there will be scheduled.
+You can cancel and experiment by commenting its line.
+
+This grid shows how to train a MusicGen EnCodec model at 32 kHz.
+"""
+
+from ._explorers import CompressionExplorer
+from ...environment import AudioCraftEnvironment
+
+
+@CompressionExplorer
+def explorer(launcher):
+    partitions = AudioCraftEnvironment.get_slurm_partitions(['team', 'global'])
+    launcher.slurm_(gpus=8, partition=partitions)
+    # use configuration for MusicGen's EnCodec model trained on monophonic audio sampled at 32 kHz
+    # MusicGen's EnCodec is trained with a total stride of 640 leading to a frame rate of 50 hz
+    launcher.bind_(solver='compression/encodec_musicgen_32khz')
+    # replace this by the desired music dataset
+    launcher.bind_(dset='internal/music_400k_32khz')
+    # launch xp
+    launcher()
+    launcher({
+        'metrics.visqol.bin': '/data/home/jadecopet/local/usr/opt/visqol',
+        'label': 'visqol',
+        'evaluate.metrics.visqol': True
+    })