test: use temp output folders and pytest.parametrize

idiap · Dec 12, 2024 · 6f34c74 · 6f34c74
1 parent 033f166
commit 6f34c74
Show file tree

Hide file tree

Showing 13 changed files with 497 additions and 536 deletions.
diff --git a/tests/data_tests/test_loader.py b/tests/data_tests/test_loader.py
diff --git a/tests/inference_tests/test_synthesize.py b/tests/inference_tests/test_synthesize.py
@@ -1,11 +1,9 @@
-import os
+from tests import run_cli
 
-from tests import get_tests_output_path, run_cli
 
-
-def test_synthesize():
+def test_synthesize(tmp_path):
     """Test synthesize.py with diffent arguments."""
-    output_path = os.path.join(get_tests_output_path(), "output.wav")
+    output_path = tmp_path / "output.wav"
     run_cli("tts --list_models")
 
     # single speaker model

diff --git a/tests/vocoder_tests/test_fullband_melgan_train.py b/tests/vocoder_tests/test_fullband_melgan_train.py
@@ -1,43 +1,42 @@
-import glob
-import os
-import shutil
-
-from tests import get_device_id, get_tests_output_path, run_cli
+from tests import get_device_id, run_cli
 from TTS.vocoder.configs import FullbandMelganConfig
 
-config_path = os.path.join(get_tests_output_path(), "test_vocoder_config.json")
-output_path = os.path.join(get_tests_output_path(), "train_outputs")
 
-config = FullbandMelganConfig(
-    batch_size=8,
-    eval_batch_size=8,
-    num_loader_workers=0,
-    num_eval_loader_workers=0,
-    run_eval=True,
-    test_delay_epochs=-1,
-    epochs=1,
-    seq_len=8192,
-    eval_split_size=1,
-    print_step=1,
-    print_eval=True,
-    data_path="tests/data/ljspeech",
-    discriminator_model_params={"base_channels": 16, "max_channels": 64, "downsample_factors": [4, 4, 4]},
-    output_path=output_path,
-)
-config.audio.do_trim_silence = True
-config.audio.trim_db = 60
-config.save_json(config_path)
+def test_train(tmp_path):
+    config_path = tmp_path / "test_vocoder_config.json"
+    output_path = tmp_path / "train_outputs"
+
+    config = FullbandMelganConfig(
+        batch_size=8,
+        eval_batch_size=8,
+        num_loader_workers=0,
+        num_eval_loader_workers=0,
+        run_eval=True,
+        test_delay_epochs=-1,
+        epochs=1,
+        seq_len=8192,
+        eval_split_size=1,
+        print_step=1,
+        print_eval=True,
+        data_path="tests/data/ljspeech",
+        discriminator_model_params={"base_channels": 16, "max_channels": 64, "downsample_factors": [4, 4, 4]},
+        output_path=output_path,
+    )
+    config.audio.do_trim_silence = True
+    config.audio.trim_db = 60
+    config.save_json(config_path)
 
-# train the model for one epoch
-command_train = f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --config_path {config_path} "
-run_cli(command_train)
+    # train the model for one epoch
+    command_train = (
+        f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --config_path {config_path} "
+    )
+    run_cli(command_train)
 
-# Find latest folder
-continue_path = max(glob.glob(os.path.join(output_path, "*/")), key=os.path.getmtime)
+    # Find latest folder
+    continue_path = max(output_path.iterdir(), key=lambda p: p.stat().st_mtime)
 
-# restore the model and continue training for one more epoch
-command_train = (
-    f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --continue_path {continue_path} "
-)
-run_cli(command_train)
-shutil.rmtree(continue_path)
+    # restore the model and continue training for one more epoch
+    command_train = (
+        f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --continue_path {continue_path} "
+    )
+    run_cli(command_train)
diff --git a/tests/vocoder_tests/test_hifigan_train.py b/tests/vocoder_tests/test_hifigan_train.py
@@ -1,43 +1,41 @@
-import glob
-import os
-import shutil
-
-from tests import get_device_id, get_tests_output_path, run_cli
+from tests import get_device_id, run_cli
 from TTS.vocoder.configs import HifiganConfig
 
-config_path = os.path.join(get_tests_output_path(), "test_vocoder_config.json")
-output_path = os.path.join(get_tests_output_path(), "train_outputs")
 
+def test_train(tmp_path):
+    config_path = tmp_path / "test_vocoder_config.json"
+    output_path = tmp_path / "train_outputs"
 
-config = HifiganConfig(
-    batch_size=8,
-    eval_batch_size=8,
-    num_loader_workers=0,
-    num_eval_loader_workers=0,
-    run_eval=True,
-    test_delay_epochs=-1,
-    epochs=1,
-    seq_len=1024,
-    eval_split_size=1,
-    print_step=1,
-    print_eval=True,
-    data_path="tests/data/ljspeech",
-    output_path=output_path,
-)
-config.audio.do_trim_silence = True
-config.audio.trim_db = 60
-config.save_json(config_path)
+    config = HifiganConfig(
+        batch_size=8,
+        eval_batch_size=8,
+        num_loader_workers=0,
+        num_eval_loader_workers=0,
+        run_eval=True,
+        test_delay_epochs=-1,
+        epochs=1,
+        seq_len=1024,
+        eval_split_size=1,
+        print_step=1,
+        print_eval=True,
+        data_path="tests/data/ljspeech",
+        output_path=output_path,
+    )
+    config.audio.do_trim_silence = True
+    config.audio.trim_db = 60
+    config.save_json(config_path)
 
-# train the model for one epoch
-command_train = f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --config_path {config_path} "
-run_cli(command_train)
+    # train the model for one epoch
+    command_train = (
+        f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --config_path {config_path} "
+    )
+    run_cli(command_train)
 
-# Find latest folder
-continue_path = max(glob.glob(os.path.join(output_path, "*/")), key=os.path.getmtime)
+    # Find latest folder
+    continue_path = max(output_path.iterdir(), key=lambda p: p.stat().st_mtime)
 
-# restore the model and continue training for one more epoch
-command_train = (
-    f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --continue_path {continue_path} "
-)
-run_cli(command_train)
-shutil.rmtree(continue_path)
+    # restore the model and continue training for one more epoch
+    command_train = (
+        f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --continue_path {continue_path} "
+    )
+    run_cli(command_train)
diff --git a/tests/vocoder_tests/test_melgan_train.py b/tests/vocoder_tests/test_melgan_train.py
@@ -1,43 +1,42 @@
-import glob
-import os
-import shutil
-
-from tests import get_device_id, get_tests_output_path, run_cli
+from tests import get_device_id, run_cli
 from TTS.vocoder.configs import MelganConfig
 
-config_path = os.path.join(get_tests_output_path(), "test_vocoder_config.json")
-output_path = os.path.join(get_tests_output_path(), "train_outputs")
 
-config = MelganConfig(
-    batch_size=4,
-    eval_batch_size=4,
-    num_loader_workers=0,
-    num_eval_loader_workers=0,
-    run_eval=True,
-    test_delay_epochs=-1,
-    epochs=1,
-    seq_len=2048,
-    eval_split_size=1,
-    print_step=1,
-    discriminator_model_params={"base_channels": 16, "max_channels": 64, "downsample_factors": [4, 4, 4]},
-    print_eval=True,
-    data_path="tests/data/ljspeech",
-    output_path=output_path,
-)
-config.audio.do_trim_silence = True
-config.audio.trim_db = 60
-config.save_json(config_path)
+def test_train(tmp_path):
+    config_path = tmp_path / "test_vocoder_config.json"
+    output_path = tmp_path / "train_outputs"
+
+    config = MelganConfig(
+        batch_size=4,
+        eval_batch_size=4,
+        num_loader_workers=0,
+        num_eval_loader_workers=0,
+        run_eval=True,
+        test_delay_epochs=-1,
+        epochs=1,
+        seq_len=2048,
+        eval_split_size=1,
+        print_step=1,
+        discriminator_model_params={"base_channels": 16, "max_channels": 64, "downsample_factors": [4, 4, 4]},
+        print_eval=True,
+        data_path="tests/data/ljspeech",
+        output_path=output_path,
+    )
+    config.audio.do_trim_silence = True
+    config.audio.trim_db = 60
+    config.save_json(config_path)
 
-# train the model for one epoch
-command_train = f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --config_path {config_path} "
-run_cli(command_train)
+    # train the model for one epoch
+    command_train = (
+        f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --config_path {config_path} "
+    )
+    run_cli(command_train)
 
-# Find latest folder
-continue_path = max(glob.glob(os.path.join(output_path, "*/")), key=os.path.getmtime)
+    # Find latest folder
+    continue_path = max(output_path.iterdir(), key=lambda p: p.stat().st_mtime)
 
-# restore the model and continue training for one more epoch
-command_train = (
-    f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --continue_path {continue_path} "
-)
-run_cli(command_train)
-shutil.rmtree(continue_path)
+    # restore the model and continue training for one more epoch
+    command_train = (
+        f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --continue_path {continue_path} "
+    )
+    run_cli(command_train)
diff --git a/tests/vocoder_tests/test_multiband_melgan_train.py b/tests/vocoder_tests/test_multiband_melgan_train.py
@@ -1,44 +1,43 @@
-import glob
-import os
-import shutil
-
-from tests import get_device_id, get_tests_output_path, run_cli
+from tests import get_device_id, run_cli
 from TTS.vocoder.configs import MultibandMelganConfig
 
-config_path = os.path.join(get_tests_output_path(), "test_vocoder_config.json")
-output_path = os.path.join(get_tests_output_path(), "train_outputs")
 
-config = MultibandMelganConfig(
-    batch_size=8,
-    eval_batch_size=8,
-    num_loader_workers=0,
-    num_eval_loader_workers=0,
-    run_eval=True,
-    test_delay_epochs=-1,
-    epochs=1,
-    seq_len=8192,
-    eval_split_size=1,
-    print_step=1,
-    print_eval=True,
-    steps_to_start_discriminator=1,
-    data_path="tests/data/ljspeech",
-    discriminator_model_params={"base_channels": 16, "max_channels": 64, "downsample_factors": [4, 4, 4]},
-    output_path=output_path,
-)
-config.audio.do_trim_silence = True
-config.audio.trim_db = 60
-config.save_json(config_path)
+def test_train(tmp_path):
+    config_path = tmp_path / "test_vocoder_config.json"
+    output_path = tmp_path / "train_outputs"
+
+    config = MultibandMelganConfig(
+        batch_size=8,
+        eval_batch_size=8,
+        num_loader_workers=0,
+        num_eval_loader_workers=0,
+        run_eval=True,
+        test_delay_epochs=-1,
+        epochs=1,
+        seq_len=8192,
+        eval_split_size=1,
+        print_step=1,
+        print_eval=True,
+        steps_to_start_discriminator=1,
+        data_path="tests/data/ljspeech",
+        discriminator_model_params={"base_channels": 16, "max_channels": 64, "downsample_factors": [4, 4, 4]},
+        output_path=output_path,
+    )
+    config.audio.do_trim_silence = True
+    config.audio.trim_db = 60
+    config.save_json(config_path)
 
-# train the model for one epoch
-command_train = f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --config_path {config_path} "
-run_cli(command_train)
+    # train the model for one epoch
+    command_train = (
+        f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --config_path {config_path} "
+    )
+    run_cli(command_train)
 
-# Find latest folder
-continue_path = max(glob.glob(os.path.join(output_path, "*/")), key=os.path.getmtime)
+    # Find latest folder
+    continue_path = max(output_path.iterdir(), key=lambda p: p.stat().st_mtime)
 
-# restore the model and continue training for one more epoch
-command_train = (
-    f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --continue_path {continue_path} "
-)
-run_cli(command_train)
-shutil.rmtree(continue_path)
+    # restore the model and continue training for one more epoch
+    command_train = (
+        f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --continue_path {continue_path} "
+    )
+    run_cli(command_train)