Skip to content

Commit 6dbb067

Browse files
sbalandisbalandi
committed
test2
1 parent dcd86f7 commit 6dbb067

10 files changed

Lines changed: 54 additions & 33 deletions

File tree

‎.github/workflows/linux.yml‎

Lines changed: 8 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -33,8 +33,6 @@ env:
3333
BASE_PRODUCT_TYPE: public_linux_ubuntu_22_04_x86_64
3434
GENAI_WHEELS_ARTIFACT_NAME: 'genai_wheels'
3535
GENAI_ARCHIVE_ARTIFACT_BASE_NAME: 'genai_archive'
36-
HF_DATASETS_CACHE: /mount/caches/pytest/datasets
37-
HF_HUB_DOWNLOAD_TIMEOUT: 60
3836

3937
jobs:
4038
smart_ci:
@@ -544,11 +542,19 @@ jobs:
544542
run_condition: ${{ fromJSON(needs.smart_ci.outputs.affected_components).RAG.test }}
545543
timeout: 30
546544
- name: 'WWB tests'
545+
env:
546+
HF_DATASETS_CACHE: /mount/caches/pytest/datasets
547+
HF_HUB_DOWNLOAD_TIMEOUT: 60
547548
cmd: 'python -m pytest -v ./tools/who_what_benchmark/tests -m "not nanollava"'
548549
run_condition: ${{ fromJSON(needs.smart_ci.outputs.affected_components).WWB.test }}
549550
timeout: 120
550551
- name: 'WWB tests (nanollava)'
552+
env:
553+
HF_DATASETS_CACHE: /mount/caches/pytest/datasets
554+
HF_HUB_DOWNLOAD_TIMEOUT: 60
551555
cmd: |
556+
echo "HF_HUB_DOWNLOAD_TIMEOUT=$HF_HUB_DOWNLOAD_TIMEOUT"
557+
echo "HF_DATASETS_CACHE=$HF_DATASETS_CACHE"
552558
python -m pip install transformers==4.48.0
553559
python -m pytest -v ./tools/who_what_benchmark/tests -m nanollava
554560
run_condition: ${{ fromJSON(needs.smart_ci.outputs.affected_components).WWB.test }}

‎.github/workflows/mac.yml‎

Lines changed: 6 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -25,8 +25,6 @@ env:
2525
OV_CACHE: ~/.cache/ov_cache/
2626
CLEANUP_CACHE: 1
2727
OPENVINO_LOG_LEVEL: 4
28-
HF_DATASETS_CACHE: ~/.cache/pytest/datasets
29-
HF_HUB_DOWNLOAD_TIMEOUT: 60
3028

3129
jobs:
3230
smart_ci:
@@ -470,10 +468,16 @@ jobs:
470468
run_condition: ${{ fromJSON(needs.smart_ci.outputs.affected_components).RAG.test }}
471469
timeout: 30
472470
- name: 'WWB tests'
471+
env:
472+
HF_DATASETS_CACHE: ~/.cache/pytest/datasets
473+
HF_HUB_DOWNLOAD_TIMEOUT: 60
473474
cmd: 'python -m pytest -v ./tools/who_what_benchmark/tests -m "not nanollava"'
474475
run_condition: ${{ fromJSON(needs.smart_ci.outputs.affected_components).WWB.test }}
475476
timeout: 120
476477
- name: 'WWB tests (nanollava)'
478+
env:
479+
HF_DATASETS_CACHE: ~/.cache/pytest/datasets
480+
HF_HUB_DOWNLOAD_TIMEOUT: 60
477481
cmd: |
478482
python -m pip install transformers==4.48.0
479483
python -m pytest -v ./tools/who_what_benchmark/tests -m nanollava

‎.github/workflows/manylinux_2_28.yml‎

Lines changed: 8 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
name: Manylinux 2_28
1+
sccachewname: Manylinux 2_28
22
on:
33
workflow_dispatch:
44
pull_request:
@@ -32,9 +32,7 @@ env:
3232
ARTIFACTS_SHARE: '/mount/build-artifacts'
3333
BASE_PRODUCT_TYPE: public_manylinux_2_28_x86_64
3434
GENAI_WHEELS_ARTIFACT_NAME: 'genai_wheels'
35-
GENAI_ARCHIVE_ARTIFACT_BASE_NAME: 'genai_archive'
36-
HF_DATASETS_CACHE: /mount/caches/pytest/datasets
37-
HF_HUB_DOWNLOAD_TIMEOUT: 60
35+
wb GENAI_ARCHIVE_ARTIFACT_BASE_NAME: 'genai_archive'
3836

3937
jobs:
4038
smart_ci:
@@ -485,10 +483,16 @@ jobs:
485483
run_condition: ${{ fromJSON(needs.smart_ci.outputs.affected_components).RAG.test }}
486484
timeout: 30
487485
- name: 'WWB tests'
486+
env:
487+
HF_DATASETS_CACHE: /mount/caches/pytest/datasets
488+
HF_HUB_DOWNLOAD_TIMEOUT: 60
488489
cmd: 'python -m pytest -v ./tools/who_what_benchmark/tests -m "not nanollava"'
489490
run_condition: ${{ fromJSON(needs.smart_ci.outputs.affected_components).WWB.test }}
490491
timeout: 120
491492
- name: 'WWB tests (nanollava)'
493+
env:
494+
HF_DATASETS_CACHE: /mount/caches/pytest/datasets
495+
HF_HUB_DOWNLOAD_TIMEOUT: 60
492496
cmd: |
493497
python -m pip install transformers==4.48.0
494498
python -m pytest -v ./tools/who_what_benchmark/tests -m nanollava

‎.github/workflows/windows.yml‎

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -631,8 +631,14 @@ jobs:
631631
run_condition: ${{ fromJSON(needs.smart_ci.outputs.affected_components).RAG.test }}
632632
timeout: 30
633633
- name: 'WWB tests'
634+
env:
635+
HF_DATASETS_CACHE: C:/mount/caches/datasets
636+
HF_HUB_DOWNLOAD_TIMEOUT: 60
634637
cmd: 'python -m pytest -s -v tools/who_what_benchmark/tests -m "not nanollava"'
635638
run_condition: ${{ fromJSON(needs.smart_ci.outputs.affected_components).WWB.test }}
639+
env:
640+
HF_DATASETS_CACHE: C:/mount/caches/datasets
641+
HF_HUB_DOWNLOAD_TIMEOUT: 60
636642
timeout: 120
637643
- name: 'WWB tests (nanollava)'
638644
cmd: |

‎tools/who_what_benchmark/tests/conftest.py‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -127,6 +127,7 @@ def run_wwb(args, env=None):
127127
stderr=subprocess.STDOUT,
128128
encoding="utf-8",
129129
env=base_env,
130+
errors="replace"
130131
)
131132
except subprocess.CalledProcessError as error:
132133
logger.error(f"'{' '.join(map(str, command))}' returned {error.returncode}. Output:\n{error.output}")

‎tools/who_what_benchmark/whowhatbench/embeddings_evaluator.py‎

Lines changed: 5 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -27,11 +27,11 @@ def prepare_default_data(num_samples=None):
2727
DATASET_NAME = "microsoft/ms_marco"
2828
NUM_SAMPLES = num_samples if num_samples else 24
2929
set_seed(42)
30-
lock = FileLock(os.path.join(lock_path, lock_file_name))
31-
with lock.acquire(timeout=300):
32-
default_dataset = datasets.load_dataset(
33-
DATASET_NAME, 'v2.1', split="test", streaming=True
34-
).shuffle(42).take(NUM_SAMPLES)
30+
# lock = FileLock(os.path.join(lock_path, lock_file_name))
31+
# with lock.acquire(timeout=300):
32+
default_dataset = datasets.load_dataset(
33+
DATASET_NAME, 'v2.1', split="test", streaming=True
34+
).shuffle(42).take(NUM_SAMPLES)
3535
return default_dataset.map(
3636
lambda x: {'passages': x['passages']['passage_text']}, remove_columns=default_dataset.column_names
3737
)

‎tools/who_what_benchmark/whowhatbench/im2im_evaluator.py‎

Lines changed: 5 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -32,11 +32,11 @@ def prepare_default_data(num_samples=None):
3232
DATASET_NAME = "paint-by-inpaint/PIPE"
3333
NUM_SAMPLES = 10 if num_samples is None else num_samples
3434
set_seed(42)
35-
lock = FileLock(os.path.join(lock_path, lock_file_name))
36-
with lock.acquire(timeout=300):
37-
default_dataset = datasets.load_dataset(
38-
DATASET_NAME, split="test", streaming=True
39-
).filter(lambda example: example["Instruction_VLM-LLM"] != "").take(NUM_SAMPLES)
35+
# lock = FileLock(os.path.join(lock_path, lock_file_name))
36+
# with lock.acquire(timeout=300):
37+
default_dataset = datasets.load_dataset(
38+
DATASET_NAME, split="test", streaming=True
39+
).filter(lambda example: example["Instruction_VLM-LLM"] != "").take(NUM_SAMPLES)
4040
return default_dataset.map(
4141
lambda x: preprocess_fn(x), remove_columns=default_dataset.column_names
4242
)

‎tools/who_what_benchmark/whowhatbench/inpaint_evaluator.py‎

Lines changed: 5 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -47,11 +47,11 @@ def prepare_default_data(num_samples=None):
4747
DATASET_NAME = "phiyodr/InpaintCOCO"
4848
NUM_SAMPLES = 10 if num_samples is None else num_samples
4949
set_seed(42)
50-
lock = FileLock(os.path.join(lock_path, lock_file_name))
51-
with lock.acquire(timeout=300):
52-
default_dataset = datasets.load_dataset(
53-
DATASET_NAME, split="test", streaming=True,
54-
).filter(lambda example: example["inpaint_caption"] != "").take(NUM_SAMPLES)
50+
# lock = FileLock(os.path.join(lock_path, lock_file_name))
51+
# with lock.acquire(timeout=300):
52+
default_dataset = datasets.load_dataset(
53+
DATASET_NAME, split="test", streaming=True,
54+
).filter(lambda example: example["inpaint_caption"] != "").take(NUM_SAMPLES)
5555
return default_dataset.map(
5656
lambda x: preprocess_fn(x), remove_columns=default_dataset.column_names
5757
)

‎tools/who_what_benchmark/whowhatbench/reranking_evaluator.py‎

Lines changed: 5 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -44,11 +44,11 @@ def prepare_default_data(num_samples=None):
4444
DATASET_NAME = "microsoft/ms_marco"
4545
NUM_SAMPLES = num_samples if num_samples else 24
4646
set_seed(42)
47-
lock = FileLock(os.path.join(lock_path, lock_file_name))
48-
with lock.acquire(timeout=300):
49-
default_dataset = datasets.load_dataset(
50-
DATASET_NAME, 'v2.1', split="test", streaming=True
51-
).shuffle(42).take(NUM_SAMPLES)
47+
# lock = FileLock(os.path.join(lock_path, lock_file_name))
48+
# with lock.acquire(timeout=300):
49+
default_dataset = datasets.load_dataset(
50+
DATASET_NAME, 'v2.1', split="test", streaming=True
51+
).shuffle(42).take(NUM_SAMPLES)
5252
return default_dataset.map(
5353
lambda x: preprocess_fn(x), remove_columns=default_dataset.column_names
5454
)

‎tools/who_what_benchmark/whowhatbench/utils.py‎

Lines changed: 5 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -181,11 +181,11 @@ def prepare_default_data_image(num_samples=None):
181181
DATASET_NAME = "ucla-contextual/contextual_test"
182182
NUM_SAMPLES = 24 if num_samples is None else num_samples
183183
set_seed(42)
184-
lock = FileLock(os.path.join(lock_path, lock_file_name))
185-
with lock.acquire(timeout=300):
186-
default_dataset = datasets.load_dataset(
187-
DATASET_NAME, split="test", streaming=True
188-
).shuffle(42).take(NUM_SAMPLES)
184+
# lock = FileLock(os.path.join(lock_path, lock_file_name))
185+
# with lock.acquire(timeout=300):
186+
default_dataset = datasets.load_dataset(
187+
DATASET_NAME, split="test", streaming=True
188+
).shuffle(42).take(NUM_SAMPLES)
189189
return default_dataset.map(
190190
lambda x: preprocess_fn(x), remove_columns=default_dataset.column_names
191191
)

0 commit comments

Comments
 (0)