spm sample exps

rwth-i6 · May 10, 2024 · 71550ac · 71550ac
1 parent 61a7a0a
commit 71550ac
Show file tree

Hide file tree

Showing 2 changed files with 28 additions and 1 deletion.
diff --git a/users/zeyer/experiments/exp2024_04_23_baselines/aed.py b/users/zeyer/experiments/exp2024_04_23_baselines/aed.py
@@ -80,6 +80,19 @@ def py():
         vocab="spm10k",
     )
 
+    train_exp(
+        "v6-bhv20-11gb-f32-bs15k-accgrad1-mgpu4-pavg100-wd1e_2-lrlin1e_5_295k-speedpertV2-spm10k-spmSample01",
+        config_11gb_v6_f32_accgrad1_mgpu4_pavg100_wd1e_4,
+        config_updates={
+            **_get_cfg_lrlin_oclr_by_bs_nep(15_000, 500),
+            "optimizer.weight_decay": 1e-2,
+            "__train_audio_preprocess": speed_pert_librosa_config,
+            "speed_pert_discrete_values": [0.7, 0.8, 0.9, 1.0, 1.1],
+        },
+        vocab="spm10k",
+        train_vocab_opts={"other_opts": {"enable_sampling": True, "alpha": 0.1}},
+    )
+
 
 # noinspection PyShadowingNames
 def train_exp(

diff --git a/users/zeyer/experiments/exp2024_04_23_baselines/ctc.py b/users/zeyer/experiments/exp2024_04_23_baselines/ctc.py
@@ -53,6 +53,19 @@ def py():
         vocab="spm10k",
     )
 
+    train_exp(
+        "v6-bhv20-11gb-f32-bs15k-accgrad1-mgpu4-pavg100-wd1e_2-lrlin1e_5_295k-speedpertV2-spm10k-spmSample01",
+        config_11gb_v6_f32_accgrad1_mgpu4_pavg100_wd1e_4,
+        config_updates={
+            **_get_cfg_lrlin_oclr_by_bs_nep(15_000, 500),
+            "optimizer.weight_decay": 1e-2,
+            "__train_audio_preprocess": speed_pert_librosa_config,
+            "speed_pert_discrete_values": [0.7, 0.8, 0.9, 1.0, 1.1],
+        },
+        vocab="spm10k",
+        train_vocab_opts={"other_opts": {"enable_sampling": True, "alpha": 0.1}},
+    )
+
 
 # noinspection PyShadowingNames
 def train_exp(
@@ -61,6 +74,7 @@ def train_exp(
     *,
     model_def: Optional[Union[ModelDefWithCfg, ModelDef[Model]]] = None,
     vocab: str = "bpe10k",
+    train_vocab_opts: Optional[Dict[str, Any]] = None,
     train_def: Optional[TrainDef[Model]] = None,
     model_config: Optional[Dict[str, Any]] = None,
     config_updates: Optional[Dict[str, Any]] = None,
@@ -82,7 +96,7 @@ def train_exp(
         _sis_setup_global_prefix()
 
     prefix = _sis_prefix + "/" + name
-    task = get_librispeech_task_raw_v2(vocab=vocab)
+    task = get_librispeech_task_raw_v2(vocab=vocab, train_vocab_opts=train_vocab_opts)
     config = config.copy()
     config = dict_update_deep(config, config_updates, config_deletes)
     if "__num_epochs" in config: