Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
chenpangpang
transformers
Commits
e26c6f03
Unverified
Commit
e26c6f03
authored
Jun 12, 2023
by
Yih-Dar
Committed by
GitHub
Jun 12, 2023
Browse files
Fix `Wav2Vec2` CI OOM (#24190)
fix Co-authored-by:
ydshieh
<
ydshieh@users.noreply.github.com
>
parent
8f093fb7
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
13 additions
and
0 deletions
+13
-0
tests/models/wav2vec2/test_modeling_tf_wav2vec2.py
tests/models/wav2vec2/test_modeling_tf_wav2vec2.py
+6
-0
tests/models/wav2vec2/test_modeling_wav2vec2.py
tests/models/wav2vec2/test_modeling_wav2vec2.py
+7
-0
No files found.
tests/models/wav2vec2/test_modeling_tf_wav2vec2.py
View file @
e26c6f03
...
...
@@ -17,6 +17,7 @@
from
__future__
import
annotations
import
copy
import
gc
import
glob
import
inspect
import
math
...
...
@@ -709,6 +710,11 @@ class TFWav2Vec2UtilsTest(unittest.TestCase):
@
require_tf
@
slow
class
TFWav2Vec2ModelIntegrationTest
(
unittest
.
TestCase
):
def
tearDown
(
self
):
super
().
tearDown
()
# clean-up as much as possible GPU memory occupied by PyTorch
gc
.
collect
()
def
_load_datasamples
(
self
,
num_samples
):
ds
=
load_dataset
(
"hf-internal-testing/librispeech_asr_dummy"
,
"clean"
,
split
=
"validation"
)
# automatic decoding with librispeech
...
...
tests/models/wav2vec2/test_modeling_wav2vec2.py
View file @
e26c6f03
...
...
@@ -14,6 +14,7 @@
# limitations under the License.
""" Testing suite for the PyTorch Wav2Vec2 model. """
import
gc
import
math
import
multiprocessing
import
os
...
...
@@ -1374,6 +1375,12 @@ class Wav2Vec2UtilsTest(unittest.TestCase):
@
require_soundfile
@
slow
class
Wav2Vec2ModelIntegrationTest
(
unittest
.
TestCase
):
def
tearDown
(
self
):
super
().
tearDown
()
# clean-up as much as possible GPU memory occupied by PyTorch
gc
.
collect
()
torch
.
cuda
.
empty_cache
()
def
_load_datasamples
(
self
,
num_samples
):
ds
=
load_dataset
(
"hf-internal-testing/librispeech_asr_dummy"
,
"clean"
,
split
=
"validation"
)
# automatic decoding with librispeech
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment