sdgym 0.12.2.dev0__tar.gz → 0.13.1.dev0__tar.gz

{sdgym-0.12.2.dev0/sdgym.egg-info → sdgym-0.13.1.dev0}/PKG-INFO

@@ -1,9 +1,9 @@
 Metadata-Version: 2.4
 Name: sdgym
-Version: 0.12.2.dev0
+Version: 0.13.1.dev0
 Summary: Benchmark tabular synthetic data generators using a variety of datasets
 Author-email: "DataCebo, Inc." <info@sdv.dev>
-License: BSL-1.1
+License-Expression: BUSL-1.1
 Project-URL: Source Code, https://github.com/sdv-dev/SDGym/
 Project-URL: Issue Tracker, https://github.com/sdv-dev/SDGym/issues
 Project-URL: Changes, https://github.com/sdv-dev/SDGym/blob/main/HISTORY.md
@@ -12,7 +12,6 @@ Project-URL: Chat, https://bit.ly/sdv-slack-invite
 Keywords: machine learning,synthetic data generation,benchmark,generative models
 Classifier: Development Status :: 2 - Pre-Alpha
 Classifier: Intended Audience :: Developers
-Classifier: License :: Free for non-commercial use
 Classifier: Natural Language :: English
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.9
@@ -20,44 +19,57 @@ Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Requires-Python: <3.14,>=3.9
+Requires-Python: <3.15,>=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: appdirs>=1.3
 Requires-Dist: boto3<2,>=1.28
 Requires-Dist: botocore<2,>=1.31
-Requires-Dist: cloudpickle>=2.1.0
+Requires-Dist: cloudpickle>=2.1.0; python_version < "3.14"
+Requires-Dist: cloudpickle>=3.1.1; python_version >= "3.14"
 Requires-Dist: compress-pickle>=1.2.0
-Requires-Dist: google-cloud-compute>=1.0.0
-Requires-Dist: google-auth>=2.0.0
+Requires-Dist: google-cloud-compute>=1.30.0
+Requires-Dist: google-auth>=2.14.1
 Requires-Dist: humanfriendly>=10.0
 Requires-Dist: numpy>=1.22.2; python_version < "3.10"
 Requires-Dist: numpy>=1.24.0; python_version >= "3.10" and python_version < "3.12"
 Requires-Dist: numpy>=1.26.0; python_version >= "3.12" and python_version < "3.13"
-Requires-Dist: numpy>=2.1.0; python_version >= "3.13"
+Requires-Dist: numpy>=2.1.0; python_version >= "3.13" and python_version < "3.14"
+Requires-Dist: numpy>=2.3.2; python_version >= "3.14"
 Requires-Dist: openpyxl>=3.1.2
-Requires-Dist: pandas<3.0.0,>=1.4.0; python_version < "3.11"
-Requires-Dist: pandas<3.0.0,>=1.5.0; python_version >= "3.11" and python_version < "3.12"
-Requires-Dist: pandas<3.0.0,>=2.1.1; python_version >= "3.12" and python_version < "3.13"
-Requires-Dist: pandas<3.0.0,>=2.2.3; python_version >= "3.13"
-Requires-Dist: psutil>=5.7
+Requires-Dist: pandas<3,>=1.4.0; python_version < "3.11"
+Requires-Dist: pandas<3,>=1.5.0; python_version >= "3.11" and python_version < "3.12"
+Requires-Dist: pandas<3,>=2.1.1; python_version >= "3.12" and python_version < "3.13"
+Requires-Dist: pandas<3,>=2.2.3; python_version >= "3.13" and python_version < "3.14"
+Requires-Dist: pandas<3,>=2.3.3; python_version >= "3.14"
+Requires-Dist: psutil>=5.8
 Requires-Dist: scikit-learn>=1.0.2; python_version < "3.10"
 Requires-Dist: scikit-learn>=1.1.0; python_version >= "3.10" and python_version < "3.11"
 Requires-Dist: scikit-learn>=1.1.3; python_version >= "3.11" and python_version < "3.12"
 Requires-Dist: scikit-learn>=1.3.1; python_version >= "3.12" and python_version < "3.13"
-Requires-Dist: scikit-learn>=1.5.2; python_version >= "3.13"
+Requires-Dist: scikit-learn>=1.5.2; python_version >= "3.13" and python_version < "3.14"
+Requires-Dist: scikit-learn>=1.8.0; python_version >= "3.14"
 Requires-Dist: scipy>=1.7.3; python_version < "3.10"
 Requires-Dist: scipy>=1.9.2; python_version >= "3.10" and python_version < "3.12"
 Requires-Dist: scipy>=1.12.0; python_version >= "3.12" and python_version < "3.13"
-Requires-Dist: scipy>=1.14.1; python_version >= "3.13"
+Requires-Dist: scipy>=1.14.1; python_version >= "3.13" and python_version < "3.14"
+Requires-Dist: scipy>=1.16.1; python_version >= "3.14"
 Requires-Dist: tabulate<0.9,>=0.8.3
-Requires-Dist: torch>=2.6.0
+Requires-Dist: torch>=1.13.0; python_version < "3.11"
+Requires-Dist: torch>=2.0.0; python_version >= "3.11" and python_version < "3.12"
+Requires-Dist: torch>=2.3.0; python_version >= "3.12" and python_version < "3.13"
+Requires-Dist: torch>=2.6.0; python_version >= "3.13" and python_version < "3.14"
+Requires-Dist: torch>=2.9.0; python_version >= "3.14"
 Requires-Dist: tqdm>=4.66.3
 Requires-Dist: XlsxWriter>=1.2.8
-Requires-Dist: rdt>=1.17.0
-Requires-Dist: sdmetrics>=0.20.1
-Requires-Dist: sdv>=1.21.0
+Requires-Dist: rdt>=1.18.2; python_version < "3.14"
+Requires-Dist: rdt>=1.20.0; python_version >= "3.14"
+Requires-Dist: sdmetrics>=0.21.0; python_version < "3.14"
+Requires-Dist: sdmetrics>=0.26.0; python_version >= "3.14"
+Requires-Dist: sdv>=1.21.0; python_version < "3.14"
+Requires-Dist: sdv>=1.33.0; python_version >= "3.14"
 Provides-Extra: dask
 Requires-Dist: dask; extra == "dask"
 Requires-Dist: distributed; extra == "dask"

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0}/pyproject.toml

@@ -5,7 +5,6 @@ authors = [{ name = 'DataCebo, Inc.', email = 'info@sdv.dev' }]
 classifiers = [
     'Development Status :: 2 - Pre-Alpha',
     'Intended Audience :: Developers',
-    'License :: Free for non-commercial use',
     'Natural Language :: English',
     'Programming Language :: Python :: 3',
     'Programming Language :: Python :: 3.9',
@@ -13,48 +12,62 @@ classifiers = [
     'Programming Language :: Python :: 3.11',
     'Programming Language :: Python :: 3.12',
     'Programming Language :: Python :: 3.13',
+    'Programming Language :: Python :: 3.14',
     'Topic :: Scientific/Engineering :: Artificial Intelligence',
 ]
 keywords = ['machine learning', 'synthetic data generation', 'benchmark', 'generative models']
 dynamic = ['version']
-license = { text = 'BSL-1.1' }
-requires-python = '>=3.9,<3.14'
+license = 'BUSL-1.1'
+license-files = ['LICENSE']
+requires-python = '>=3.9,<3.15'
 readme = 'README.md'
 dependencies = [
     'appdirs>=1.3',
     'boto3>=1.28,<2',
     'botocore>=1.31,<2',
-    'cloudpickle>=2.1.0',
+    "cloudpickle>=2.1.0;python_version<'3.14'",
+    "cloudpickle>=3.1.1;python_version>='3.14'",
     'compress-pickle>=1.2.0',
-    'google-cloud-compute>=1.0.0',
-    'google-auth>=2.0.0',
+    'google-cloud-compute>=1.30.0',
+    'google-auth>=2.14.1',
     'humanfriendly>=10.0',
     "numpy>=1.22.2;python_version<'3.10'",
     "numpy>=1.24.0;python_version>='3.10' and python_version<'3.12'",
     "numpy>=1.26.0;python_version>='3.12' and python_version<'3.13'",
-    "numpy>=2.1.0;python_version>='3.13'",
+    "numpy>=2.1.0;python_version>='3.13' and python_version<'3.14'",
+    "numpy>=2.3.2;python_version>='3.14'",
     'openpyxl>=3.1.2',
-    "pandas>=1.4.0,<3.0.0;python_version<'3.11'",
-    "pandas>=1.5.0,<3.0.0;python_version>='3.11' and python_version<'3.12'",
-    "pandas>=2.1.1,<3.0.0;python_version>='3.12' and python_version<'3.13'",
-    "pandas>=2.2.3,<3.0.0;python_version>='3.13'",
-    'psutil>=5.7',
+    "pandas>=1.4.0,<3;python_version<'3.11'",
+    "pandas>=1.5.0,<3;python_version>='3.11' and python_version<'3.12'",
+    "pandas>=2.1.1,<3;python_version>='3.12' and python_version<'3.13'",
+    "pandas>=2.2.3,<3;python_version>='3.13' and python_version<'3.14'",
+    "pandas>=2.3.3,<3;python_version>='3.14'",
+    'psutil>=5.8',
     "scikit-learn>=1.0.2;python_version<'3.10'",
     "scikit-learn>=1.1.0;python_version>='3.10' and python_version<'3.11'",
     "scikit-learn>=1.1.3;python_version>='3.11' and python_version<'3.12'",
     "scikit-learn>=1.3.1;python_version>='3.12' and python_version<'3.13'",
-    "scikit-learn>=1.5.2;python_version>='3.13'",
+    "scikit-learn>=1.5.2;python_version>='3.13' and python_version<'3.14'",
+    "scikit-learn>=1.8.0;python_version>='3.14'",
     "scipy>=1.7.3;python_version<'3.10'",
     "scipy>=1.9.2;python_version>='3.10' and python_version<'3.12'",
     "scipy>=1.12.0;python_version>='3.12' and python_version<'3.13'",
-    "scipy>=1.14.1;python_version>='3.13'",
+    "scipy>=1.14.1;python_version>='3.13' and python_version<'3.14'",
+    "scipy>=1.16.1;python_version>='3.14'",
     'tabulate>=0.8.3,<0.9',
-    "torch>=2.6.0",
+    "torch>=1.13.0;python_version<'3.11'",
+    "torch>=2.0.0;python_version>='3.11' and python_version<'3.12'",
+    "torch>=2.3.0;python_version>='3.12' and python_version<'3.13'",
+    "torch>=2.6.0;python_version>='3.13' and python_version<'3.14'",
+    "torch>=2.9.0;python_version>='3.14'",
     'tqdm>=4.66.3',
     'XlsxWriter>=1.2.8',
-    'rdt>=1.17.0',
-    'sdmetrics>=0.20.1',
-    'sdv>=1.21.0',
+    "rdt>=1.18.2;python_version<'3.14'",
+    "rdt>=1.20.0;python_version>='3.14'",
+    "sdmetrics>=0.21.0;python_version<'3.14'",
+    "sdmetrics>=0.26.0;python_version>='3.14'",
+    "sdv>=1.21.0;python_version<'3.14'",
+    "sdv>=1.33.0;python_version>='3.14'",
 ]
 
 [project.urls]
@@ -113,7 +126,6 @@ all = [
 
 [tool.setuptools]
 include-package-data = true
-license-files = ['LICENSE']
 
 [tool.setuptools.packages.find]
 include = ['sdgym', 'sdgym.*']
@@ -149,7 +161,7 @@ namespaces = false
 version = {attr = 'sdgym.__version__'}
 
 [tool.bumpversion]
-current_version = "0.12.2.dev0"
+current_version = "0.13.1.dev0"
 parse = '(?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(\.(?P<release>[a-z]+)(?P<candidate>\d+))?'
 serialize = [
     '{major}.{minor}.{patch}.{release}{candidate}',

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0}/sdgym/__init__.py

@@ -8,7 +8,7 @@ __author__ = 'DataCebo, Inc.'
 __copyright__ = 'Copyright (c) 2022 DataCebo, Inc.'
 __email__ = 'info@sdv.dev'
 __license__ = 'BSL-1.1'
-__version__ = '0.12.2.dev0'
+__version__ = '0.13.1.dev0'
 
 import logging
 

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0}/sdgym/_benchmark/credentials_utils.py

@@ -74,7 +74,7 @@ def sdv_install_cmd(credentials):
 pip install sdv-installer
 
 python -c "from sdv_installer.installation.installer import install_packages; \\
-install_packages(username='{username}', license_key='{license_key}', package='sdv-enterprise')"
+install_packages(username='{username}', license_key='{license_key}')"
 """)
 
 

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0}/sdgym/benchmark.py

@@ -514,7 +514,7 @@ def _compute_scores(
         for metric_name, metric in metrics.items():
             scores.append({
                 'metric': metric_name,
-                'error': 'Metric Timeout',
+                'Error': 'Metric Timeout',
             })
             # re-inject list to multiprocessing output
             output['scores'] = scores
@@ -537,7 +537,7 @@ def _compute_scores(
             scores[-1].update({
                 'score': score,
                 'normalized_score': normalized_score,
-                'error': error,
+                'Error': error,
                 'metric_time': calculate_score_time(start),
             })
             # re-inject list to multiprocessing output
@@ -603,7 +603,7 @@ def _score(
         output = {}
 
     output['timeout'] = True  # To be deleted if there is no error
-    output['error'] = 'Load Timeout'  # To be deleted if there is no error
+    output['Error'] = 'Load Timeout'  # To be deleted if there is no error
     try:
         LOGGER.info(
             'Running %s on %s dataset %s; %s',
@@ -615,7 +615,7 @@ def _score(
 
         output['dataset_size'] = get_size_of(data) / N_BYTES_IN_MB
         # To be deleted if there is no error
-        output['error'] = 'Synthesizer Timeout'
+        output['Error'] = 'Synthesizer Timeout'
 
         try:
             synthetic_data, train_time, sample_time, synthesizer_size, peak_memory = _synthesize(
@@ -642,7 +642,7 @@ def _score(
             )
 
             # No error so far. _compute_scores tracks its own errors by metric
-            del output['error']
+            del output['Error']
             _compute_scores(
                 metrics,
                 data,
@@ -671,14 +671,14 @@ def _score(
             output['peak_memory'] = err.peak_memory
 
             output['exception'] = err.exception
-            output['error'] = err.error
+            output['Error'] = err.error
             output['timeout'] = False
 
     except Exception:
         LOGGER.exception('Error running %s on dataset %s;', synthesizer['name'], dataset_name)
         exception, error = format_exception()
         output['exception'] = exception
-        output['error'] = error
+        output['Error'] = error
         output['timeout'] = False  # There was no timeout
 
     finally:
@@ -744,7 +744,7 @@ def _score_with_timeout(
         thread.join(timeout)
         if thread.is_alive():
             LOGGER.error('Timeout running %s on dataset %s;', synthesizer['name'], dataset_name)
-            return {'timeout': True, 'error': 'Synthesizer Timeout'}
+            return {'timeout': True, 'Error': 'Synthesizer Timeout'}
 
         return output
 
@@ -815,8 +815,8 @@ def _format_output(
     for score in output.get('scores', []):
         scores.insert(len(scores.columns), score['metric'], score['normalized_score'])
 
-    if 'error' in output:
-        scores['error'] = output['error']
+    if 'Error' in output:
+        scores['Error'] = output['Error']
 
     return scores
 
@@ -1085,8 +1085,8 @@ def _add_adjusted_scores(scores, timeout):
 
         fit_times = scores.loc[dataset_mask, 'Train_Time'].fillna(0)
         sample_times = scores.loc[dataset_mask, 'Sample_Time'].fillna(0)
-        if 'error' in scores.columns:
-            errors = scores.loc[dataset_mask, 'error']
+        if 'Error' in scores.columns:
+            errors = scores.loc[dataset_mask, 'Error']
         else:
             errors = pd.Series([None] * dataset_mask.sum(), index=scores.index[dataset_mask])
 

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0}/sdgym/cli/__main__.py

@@ -39,13 +39,13 @@ def _print_table(data, sort=None, reverse=False, format=None):
         for field, formatter in format.items():
             data[field] = data[field].apply(formatter)
 
-    if 'error' in data:
-        error = data['error']
+    if 'Error' in data:
+        error = data['Error']
         if pd.isna(error).all():
-            del data['error']
+            del data['Error']
         else:
             long_error = error.str.len() > 30
-            data.loc[long_error, 'error'] = error[long_error].str[:30] + '...'
+            data.loc[long_error, 'Error'] = error[long_error].str[:30] + '...'
 
     print(tabulate.tabulate(data, tablefmt='github', headers=data.columns, showindex=False))  # noqa: T201
 

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0}/sdgym/cli/summary.py

@@ -35,11 +35,11 @@ def preprocess(data):
     bydataset = grouped.mean()
     data = bydataset.reset_index()
 
-    if 'error' in data.columns:
+    if 'Error' in data.columns:
         errors = data.error.fillna('')
         for message, column in KNOWN_ERRORS:
             data[column] = errors.str.contains(message)
-            data.loc[data[column], 'error'] = np.nan
+            data.loc[data[column], 'Error'] = np.nan
 
     return data
 
@@ -122,7 +122,7 @@ def summarize(data, baselines=(), datasets=None):
         baseline_scores = baseline_data.set_index('Dataset').Quality_Score
         results[f'beat_{baseline.lower()}'] = _beat_baseline(data, baseline_scores)
 
-    if 'error' in data.columns:
+    if 'Error' in data.columns:
         grouped = data.groupby('Synthesizer')
         for _, error_column in KNOWN_ERRORS:
             results[error_column] = grouped[error_column].sum()
@@ -135,7 +135,7 @@ def summarize(data, baselines=(), datasets=None):
 
 
 def _error_counts(data):
-    if 'error' in data.columns:
+    if 'Error' in data.columns:
         return data.error.value_counts()
     return 0
 
@@ -158,8 +158,8 @@ def errors_summary(data):
     Returns:
         pandas.DataFrame
     """
-    if 'error' in data.columns:
-        all_errors = pd.DataFrame(_error_counts(data)).rename(columns={'error': 'all'})
+    if 'Error' in data.columns:
+        all_errors = pd.DataFrame(_error_counts(data)).rename(columns={'Error': 'all'})
         synthesizer_errors = data.groupby('Synthesizer').apply(_error_counts).pivot_table(level=0)
         for synthesizer, errors in synthesizer_errors.items():
             all_errors[synthesizer] = errors.fillna(0).astype(int)

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0}/sdgym/result_writer.py

@@ -12,6 +12,7 @@ import yaml
 from openpyxl import load_workbook
 
 from sdgym.s3 import parse_s3_path
+from sdgym.utils import _set_column_width
 
 
 class ResultsWriter(ABC):
@@ -79,6 +80,7 @@ class LocalResultsWriter:
         with writer:
             for sheet_name, df in data.items():
                 df.to_excel(writer, sheet_name=sheet_name, index=index)
+                _set_column_width(writer, df, sheet_name)
 
         wb = load_workbook(file_path)
         for sheet_name in reversed(data.keys()):

sdgym-0.13.1.dev0/sdgym/run_benchmark/run_benchmark.py

@@ -0,0 +1,206 @@
+"""Script to run a benchmark and upload results to S3."""
+
+import json
+import os
+from datetime import datetime, timezone
+
+from botocore.exceptions import ClientError
+
+from sdgym._benchmark.benchmark import (
+    _benchmark_multi_table_compute_gcp,
+    _benchmark_single_table_compute_gcp,
+)
+from sdgym.run_benchmark.utils import (
+    KEY_DATE_FILE,
+    OUTPUT_DESTINATION_AWS,
+    _exclude_datasets,
+    _parse_args,
+    get_result_folder_name,
+    post_benchmark_launch_message,
+)
+from sdgym.s3 import get_s3_client, parse_s3_path
+
+SINGLE_TABLE_DATASETS = [
+    'adult',
+    'alarm',
+    'census',
+    'child',
+    'covtype',
+    'expedia_hotel_logs',
+    'insurance',
+    'intrusion',
+    'news',
+]
+MULTI_TABLE_DATASETS = [
+    'WebKP',
+    'DCG',
+    'UW_std',
+    'Same_gen',
+    'CORA',
+    'got_families',
+    'SalesDB',
+    'UTube',
+    'Student_loan',
+    'Hepatitis_std',
+    'Elti',
+    'Bupa',
+    'Toxicology',
+    'imdb_ijs',
+    'ftp',
+    'imdb_small',
+    'imdb_MovieLens',
+    'Pima',
+    'university',
+    'legalActs',
+    'Dunur',
+    'Mesh',
+    'world',
+    'airbnb-simplified',
+    'trains',
+    'FNHK',
+    'fake_hotels',
+    'SAT',
+    'genes',
+    'Biodegradability',
+    'Pyrimidine',
+    'mutagenesis',
+    'restbase',
+    'Triazine',
+    'Carcinogenesis',
+    'fake_hotels_extended',
+    'Mooney_Family',
+    'PTE',
+    'Facebook',
+    'multi_table_ID_demo_dataset',
+    'SAP',
+    'Chess',
+    'Countries',
+    'NCAA',
+    'Atherosclerosis',
+    'nations',
+    'TubePricing',
+    'financial',
+    'Accidents',
+    'MuskSmall',
+    'NBA',
+    'AustralianFootball',
+    'PremierLeague',
+    'OMOP_CDM_dayz',
+]
+
+
+def _get_benchmark_setup(modality):
+    """Get the benchmark setup for a given modality.
+
+    The setup includes the method to run the benchmark and the job split,
+    which is a list of tuples where each tuple contains a list of synthesizers and
+    a list of datasets to run those synthesizers on.
+    """
+    if modality == 'single_table':
+        real_tab_former_to_exclude = ['covtype', 'intrusion', 'expedia_hotel_logs', 'census']
+        gan_to_exclude = ['covtype', 'intrusion']
+        job_split = [
+            (['ColumnSynthesizer', 'GaussianCopulaSynthesizer'], SINGLE_TABLE_DATASETS),
+            (['TVAESynthesizer'], SINGLE_TABLE_DATASETS),
+            (['SegmentSynthesizer'], SINGLE_TABLE_DATASETS),
+            (['XGCSynthesizer'], SINGLE_TABLE_DATASETS),
+            (['BootstrapSynthesizer'], SINGLE_TABLE_DATASETS),
+            (['CTGANSynthesizer'], _exclude_datasets(SINGLE_TABLE_DATASETS, gan_to_exclude)),
+            (['CopulaGANSynthesizer'], _exclude_datasets(SINGLE_TABLE_DATASETS, gan_to_exclude)),
+            (
+                ['RealTabFormerSynthesizer'],
+                _exclude_datasets(SINGLE_TABLE_DATASETS, real_tab_former_to_exclude),
+            ),
+        ]
+        for dataset in real_tab_former_to_exclude:
+            job_split.append((['RealTabFormerSynthesizer'], [dataset]))
+
+        for dataset in gan_to_exclude:
+            job_split.append((['CTGANSynthesizer'], [dataset]))
+            job_split.append((['CopulaGANSynthesizer'], [dataset]))
+
+        return {
+            'method': _benchmark_single_table_compute_gcp,
+            'job_split': job_split,
+        }
+
+    if modality == 'multi_table':
+        hma_to_exclude = [
+            'Accidents',
+            'AustralianFootball',
+            'Countries',
+            'MuskSmall',
+            'NBA',
+            'OMOP_CDM_dayz',
+            'PremierLeague',
+            'SalesDB',
+            'airbnb-simplified',
+            'imdb_ijs',
+            'legalActs',
+            'SAP',
+            'imdb_MovieLens',
+        ]
+        job_split = [
+            (['HSASynthesizer', 'IndependentSynthesizer'], MULTI_TABLE_DATASETS),
+            (['HMASynthesizer'], _exclude_datasets(MULTI_TABLE_DATASETS, hma_to_exclude)),
+        ]
+        for dataset in hma_to_exclude:
+            job_split.append((['HMASynthesizer'], [dataset]))
+
+        return {
+            'method': _benchmark_multi_table_compute_gcp,
+            'job_split': job_split,
+        }
+
+
+def append_benchmark_run(
+    aws_access_key_id, aws_secret_access_key, date_str, modality='single_table'
+):
+    """Append a new benchmark run to the benchmark dates file in S3."""
+    s3_client = get_s3_client(
+        aws_access_key_id=aws_access_key_id,
+        aws_secret_access_key=aws_secret_access_key,
+    )
+    bucket, prefix = parse_s3_path(OUTPUT_DESTINATION_AWS)
+    try:
+        object = s3_client.get_object(Bucket=bucket, Key=f'{prefix}{modality}/{KEY_DATE_FILE}')
+        body = object['Body'].read().decode('utf-8')
+        data = json.loads(body)
+    except ClientError as e:
+        if e.response['Error']['Code'] == 'NoSuchKey':
+            data = {'runs': []}
+        else:
+            raise RuntimeError(f'Failed to read {KEY_DATE_FILE} from S3: {e}')
+
+    data['runs'].append({'date': date_str, 'folder_name': get_result_folder_name(date_str)})
+    data['runs'] = sorted(data['runs'], key=lambda x: x['date'])
+    s3_client.put_object(
+        Bucket=bucket,
+        Key=f'{prefix}{modality}/{KEY_DATE_FILE}',
+        Body=json.dumps(data).encode('utf-8'),
+    )
+
+
+def main():
+    """Main function to run the benchmark and upload results."""
+    args = _parse_args()
+    aws_access_key_id = os.getenv('AWS_ACCESS_KEY_ID')
+    aws_secret_access_key = os.getenv('AWS_SECRET_ACCESS_KEY')
+    date_str = datetime.now(timezone.utc).strftime('%Y-%m-%d')
+    modality = args.modality
+    benchmark_setup = _get_benchmark_setup(modality)
+    for synthesizers, datasets in benchmark_setup['job_split']:
+        benchmark_setup['method'](
+            output_destination=OUTPUT_DESTINATION_AWS,
+            credential_filepath=os.getenv('CREDENTIALS_FILEPATH'),
+            synthesizers=synthesizers,
+            sdv_datasets=datasets,
+            timeout=345600,  # 4 days
+        )
+
+    append_benchmark_run(aws_access_key_id, aws_secret_access_key, date_str, modality=modality)
+    post_benchmark_launch_message(date_str, compute_service='GCP', modality=modality)
+
+
+if __name__ == '__main__':
+    main()

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0}/sdgym/run_benchmark/upload_benchmark_results.py

@@ -29,6 +29,7 @@ from sdgym.run_benchmark.utils import (
     get_df_to_plot,
 )
 from sdgym.s3 import S3_REGION, parse_s3_path
+from sdgym.utils import _set_column_width
 
 LOGGER = logging.getLogger(__name__)
 SYNTHESIZER_TO_GLOBAL_POSITION = {
@@ -231,7 +232,6 @@ def get_model_details(summary, results, df_to_plot, modality):
     with open(SYNTHESIZER_DESCRIPTION_PATH, 'r', encoding='utf-8') as f:
         synthesizer_info = yaml.safe_load(f) or {}
 
-    err_column = 'error' if 'error' in results.columns else 'Error'
     paretos_synthesizers = (
         df_to_plot.loc[df_to_plot['Pareto'].eq(True), 'Synthesizer'].astype(str).add('Synthesizer')
     )
@@ -258,18 +258,23 @@ def get_model_details(summary, results, df_to_plot, modality):
     model_details['Number of datasets - Wins'] = (
         model_details['Synthesizer'].map(wins).fillna(0).astype(int)
     )
-    timeout_counts = (
-        results
-        .loc[results[err_column].eq('Synthesizer Timeout')]
-        .groupby('Synthesizer')['Dataset']
-        .nunique()
-    )
-    error_counts = (
-        results
-        .loc[results[err_column].notna() & ~results[err_column].eq('Synthesizer Timeout')]
-        .groupby('Synthesizer')['Dataset']
-        .nunique()
-    )
+    if 'Error' in results.columns:
+        timeout_counts = (
+            results
+            .loc[results['Error'].eq('Synthesizer Timeout')]
+            .groupby('Synthesizer')['Dataset']
+            .nunique()
+        )
+        error_counts = (
+            results
+            .loc[results['Error'].notna() & ~results['Error'].eq('Synthesizer Timeout')]
+            .groupby('Synthesizer')['Dataset']
+            .nunique()
+        )
+    else:
+        timeout_counts = pd.Series(0, index=model_details['Synthesizer'])
+        error_counts = pd.Series(0, index=model_details['Synthesizer'])
+
     model_details['Number of datasets - Timeout'] = (
         model_details['Synthesizer'].map(timeout_counts).fillna(0).astype(int)
     )
@@ -313,7 +318,8 @@ def update_table_aws(s3_client, bucket, filename, table, reference_column):
     updated_table = pd.concat([existing_table, table], ignore_index=True)
     output = io.BytesIO()
     with pd.ExcelWriter(output, engine='openpyxl') as writer:
-        updated_table.to_excel(writer, index=False)
+        updated_table.to_excel(writer, index=False, sheet_name='Sheet1')
+        _set_column_width(writer, updated_table, 'Sheet1')
 
     output.seek(0)
     s3_client.upload_fileobj(output, bucket, filename)

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0}/sdgym/run_benchmark/utils.py

@@ -6,13 +6,15 @@ from datetime import datetime
 from urllib.parse import parse_qs, quote_plus, urlparse
 
 import numpy as np
+import pandas as pd
+from scipy.interpolate import interp1d
 from slack_sdk import WebClient
 
 from sdgym.s3 import parse_s3_path
 
 OUTPUT_DESTINATION_AWS = 's3://sdgym-benchmark/Benchmarks/'
 DEBUG_SLACK_CHANNEL = 'sdv-alerts-debug'
-SLACK_CHANNEL = 'sdv-alerts'
+SLACK_CHANNEL = 'sdgym'
 KEY_DATE_FILE = '_BENCHMARK_DATES.json'
 PLOTLY_MARKERS = [
     'circle',
@@ -45,18 +47,7 @@ PLOTLY_MARKERS = [
     'diamond-cross',
     'diamond-x',
 ]
-
-# The synthesizers inside the same list will be run by the same ec2 instance
-SYNTHESIZERS_SPLIT_SINGLE_TABLE = [
-    ['UniformSynthesizer', 'ColumnSynthesizer', 'GaussianCopulaSynthesizer', 'TVAESynthesizer'],
-    ['CopulaGANSynthesizer'],
-    ['CTGANSynthesizer'],
-    ['RealTabFormerSynthesizer'],
-]
-SYNTHESIZERS_SPLIT_MULTI_TABLE = [
-    ['HMASynthesizer'],
-    ['HSASynthesizer', 'IndependentSynthesizer', 'MultiTableUniformSynthesizer'],
-]
+PLOT_PADDING = 0.25
 
 
 def _get_filename_to_gdrive_link():
@@ -104,7 +95,7 @@ def post_slack_message(channel, text):
 
 
 def post_benchmark_launch_message(date_str, compute_service='AWS', modality='single_table'):
-    """Post a message to the SDV Alerts Slack channel when the benchmark is launched."""
+    """Post a message to the sdgym Slack channel when the benchmark is launched."""
     channel = SLACK_CHANNEL
     folder_name = get_result_folder_name(date_str)
     bucket, prefix = parse_s3_path(OUTPUT_DESTINATION_AWS)
@@ -116,7 +107,7 @@ def post_benchmark_launch_message(date_str, compute_service='AWS', modality='sin
 
 
 def post_benchmark_uploaded_message(folder_name, commit_url=None, modality='single_table'):
-    """Post benchmark uploaded message to sdv-alerts slack channel."""
+    """Post benchmark uploaded message to the sdgym Slack channel."""
     file_to_gdrive_link = _get_filename_to_gdrive_link()
     channel = SLACK_CHANNEL
     bucket, prefix = parse_s3_path(OUTPUT_DESTINATION_AWS)
@@ -136,6 +127,34 @@ def post_benchmark_uploaded_message(folder_name, commit_url=None, modality='sing
     post_slack_message(channel, body)
 
 
+def _add_pareto_curve_extremity_points(df_to_plot):
+    """Add extremity points to the Pareto curve for better visualization."""
+    pareto = df_to_plot.loc[df_to_plot['Pareto']].sort_values('Aggregated_Time')
+    if len(pareto) < 2:
+        return df_to_plot.reset_index(drop=True)  # Not enough points to define a curve
+
+    interp = interp1d(
+        pareto['Log10 Aggregated_Time'],
+        pareto['Quality_Score'],
+        kind='linear',
+        fill_value='extrapolate',
+    )
+
+    min_log = np.log10(df_to_plot['Aggregated_Time'].min()) - PLOT_PADDING
+    max_log = np.log10(df_to_plot['Aggregated_Time'].max()) + PLOT_PADDING
+    extremities = pd.DataFrame({
+        'Synthesizer': np.nan,
+        'Aggregated_Time': 10 ** np.array([min_log, max_log]),
+        'Quality_Score': interp([min_log, max_log]),
+        'Log10 Aggregated_Time': [min_log, max_log],
+        'Pareto': True,
+        'Color': '#01E0C9',
+        'Marker': np.nan,
+    })
+
+    return pd.concat([df_to_plot, extremities], ignore_index=True).reset_index(drop=True)
+
+
 def get_df_to_plot(benchmark_result):
     """Get the data to plot from the benchmark result.
 
@@ -177,8 +196,9 @@ def get_df_to_plot(benchmark_result):
     }
     df_to_plot['Marker'] = df_to_plot['Synthesizer'].map(marker_map)
     df_to_plot = df_to_plot.rename(columns={'Adjusted_Quality_Score': 'Quality_Score'})
+    df_to_plot = df_to_plot.drop(columns=['Cumulative Quality Score'])
 
-    return df_to_plot.drop(columns=['Cumulative Quality Score']).reset_index(drop=True)
+    return _add_pareto_curve_extremity_points(df_to_plot)
 
 
 def _parse_args():
@@ -203,3 +223,8 @@ def _extract_google_file_id(google_drive_link):
             return parsed.path.split(marker, 1)[1].split('/', 1)[0]
 
     raise ValueError(f'Invalid Google Drive link format: {google_drive_link}')
+
+
+def _exclude_datasets(datasets, dataset_to_exclude):
+    """Exclude datasets that are in the dataset_to_exclude list."""
+    return [dataset for dataset in datasets if dataset not in dataset_to_exclude]

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0}/sdgym/utils.py

@@ -11,6 +11,7 @@ import humanfriendly
 import numpy as np
 import pandas as pd
 import psutil
+from openpyxl.utils import get_column_letter
 
 from sdgym.errors import SDGymError
 from sdgym.synthesizers.base import BaselineSynthesizer
@@ -195,3 +196,11 @@ def convert_metadata_to_sdmetrics(metadata_dict):
     """Convert a sdv metadata dictionary into sdmetrics expected metadata."""
     table_name = next(iter(metadata_dict['tables']))
     return metadata_dict['tables'][table_name]
+
+
+def _set_column_width(writer, df, sheet_name):
+    worksheet = writer.sheets[sheet_name]
+    for col_idx, column in enumerate(df.columns, 1):
+        max_length = max(df[column].astype(str).map(len).max(), len(column))
+        column_letter = get_column_letter(col_idx)
+        worksheet.column_dimensions[column_letter].width = max_length + 2

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0/sdgym.egg-info}/PKG-INFO

@@ -1,9 +1,9 @@
 Metadata-Version: 2.4
 Name: sdgym
-Version: 0.12.2.dev0
+Version: 0.13.1.dev0
 Summary: Benchmark tabular synthetic data generators using a variety of datasets
 Author-email: "DataCebo, Inc." <info@sdv.dev>
-License: BSL-1.1
+License-Expression: BUSL-1.1
 Project-URL: Source Code, https://github.com/sdv-dev/SDGym/
 Project-URL: Issue Tracker, https://github.com/sdv-dev/SDGym/issues
 Project-URL: Changes, https://github.com/sdv-dev/SDGym/blob/main/HISTORY.md
@@ -12,7 +12,6 @@ Project-URL: Chat, https://bit.ly/sdv-slack-invite
 Keywords: machine learning,synthetic data generation,benchmark,generative models
 Classifier: Development Status :: 2 - Pre-Alpha
 Classifier: Intended Audience :: Developers
-Classifier: License :: Free for non-commercial use
 Classifier: Natural Language :: English
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.9
@@ -20,44 +19,57 @@ Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Requires-Python: <3.14,>=3.9
+Requires-Python: <3.15,>=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: appdirs>=1.3
 Requires-Dist: boto3<2,>=1.28
 Requires-Dist: botocore<2,>=1.31
-Requires-Dist: cloudpickle>=2.1.0
+Requires-Dist: cloudpickle>=2.1.0; python_version < "3.14"
+Requires-Dist: cloudpickle>=3.1.1; python_version >= "3.14"
 Requires-Dist: compress-pickle>=1.2.0
-Requires-Dist: google-cloud-compute>=1.0.0
-Requires-Dist: google-auth>=2.0.0
+Requires-Dist: google-cloud-compute>=1.30.0
+Requires-Dist: google-auth>=2.14.1
 Requires-Dist: humanfriendly>=10.0
 Requires-Dist: numpy>=1.22.2; python_version < "3.10"
 Requires-Dist: numpy>=1.24.0; python_version >= "3.10" and python_version < "3.12"
 Requires-Dist: numpy>=1.26.0; python_version >= "3.12" and python_version < "3.13"
-Requires-Dist: numpy>=2.1.0; python_version >= "3.13"
+Requires-Dist: numpy>=2.1.0; python_version >= "3.13" and python_version < "3.14"
+Requires-Dist: numpy>=2.3.2; python_version >= "3.14"
 Requires-Dist: openpyxl>=3.1.2
-Requires-Dist: pandas<3.0.0,>=1.4.0; python_version < "3.11"
-Requires-Dist: pandas<3.0.0,>=1.5.0; python_version >= "3.11" and python_version < "3.12"
-Requires-Dist: pandas<3.0.0,>=2.1.1; python_version >= "3.12" and python_version < "3.13"
-Requires-Dist: pandas<3.0.0,>=2.2.3; python_version >= "3.13"
-Requires-Dist: psutil>=5.7
+Requires-Dist: pandas<3,>=1.4.0; python_version < "3.11"
+Requires-Dist: pandas<3,>=1.5.0; python_version >= "3.11" and python_version < "3.12"
+Requires-Dist: pandas<3,>=2.1.1; python_version >= "3.12" and python_version < "3.13"
+Requires-Dist: pandas<3,>=2.2.3; python_version >= "3.13" and python_version < "3.14"
+Requires-Dist: pandas<3,>=2.3.3; python_version >= "3.14"
+Requires-Dist: psutil>=5.8
 Requires-Dist: scikit-learn>=1.0.2; python_version < "3.10"
 Requires-Dist: scikit-learn>=1.1.0; python_version >= "3.10" and python_version < "3.11"
 Requires-Dist: scikit-learn>=1.1.3; python_version >= "3.11" and python_version < "3.12"
 Requires-Dist: scikit-learn>=1.3.1; python_version >= "3.12" and python_version < "3.13"
-Requires-Dist: scikit-learn>=1.5.2; python_version >= "3.13"
+Requires-Dist: scikit-learn>=1.5.2; python_version >= "3.13" and python_version < "3.14"
+Requires-Dist: scikit-learn>=1.8.0; python_version >= "3.14"
 Requires-Dist: scipy>=1.7.3; python_version < "3.10"
 Requires-Dist: scipy>=1.9.2; python_version >= "3.10" and python_version < "3.12"
 Requires-Dist: scipy>=1.12.0; python_version >= "3.12" and python_version < "3.13"
-Requires-Dist: scipy>=1.14.1; python_version >= "3.13"
+Requires-Dist: scipy>=1.14.1; python_version >= "3.13" and python_version < "3.14"
+Requires-Dist: scipy>=1.16.1; python_version >= "3.14"
 Requires-Dist: tabulate<0.9,>=0.8.3
-Requires-Dist: torch>=2.6.0
+Requires-Dist: torch>=1.13.0; python_version < "3.11"
+Requires-Dist: torch>=2.0.0; python_version >= "3.11" and python_version < "3.12"
+Requires-Dist: torch>=2.3.0; python_version >= "3.12" and python_version < "3.13"
+Requires-Dist: torch>=2.6.0; python_version >= "3.13" and python_version < "3.14"
+Requires-Dist: torch>=2.9.0; python_version >= "3.14"
 Requires-Dist: tqdm>=4.66.3
 Requires-Dist: XlsxWriter>=1.2.8
-Requires-Dist: rdt>=1.17.0
-Requires-Dist: sdmetrics>=0.20.1
-Requires-Dist: sdv>=1.21.0
+Requires-Dist: rdt>=1.18.2; python_version < "3.14"
+Requires-Dist: rdt>=1.20.0; python_version >= "3.14"
+Requires-Dist: sdmetrics>=0.21.0; python_version < "3.14"
+Requires-Dist: sdmetrics>=0.26.0; python_version >= "3.14"
+Requires-Dist: sdv>=1.21.0; python_version < "3.14"
+Requires-Dist: sdv>=1.33.0; python_version >= "3.14"
 Provides-Extra: dask
 Requires-Dist: dask; extra == "dask"
 Requires-Dist: distributed; extra == "dask"

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0}/sdgym.egg-info/requires.txt

@@ -1,20 +1,15 @@
 appdirs>=1.3
 boto3<2,>=1.28
 botocore<2,>=1.31
-cloudpickle>=2.1.0
 compress-pickle>=1.2.0
-google-cloud-compute>=1.0.0
-google-auth>=2.0.0
+google-cloud-compute>=1.30.0
+google-auth>=2.14.1
 humanfriendly>=10.0
 openpyxl>=3.1.2
-psutil>=5.7
+psutil>=5.8
 tabulate<0.9,>=0.8.3
-torch>=2.6.0
 tqdm>=4.66.3
 XlsxWriter>=1.2.8
-rdt>=1.17.0
-sdmetrics>=0.20.1
-sdv>=1.21.0
 
 [:python_version < "3.10"]
 numpy>=1.22.2
@@ -22,7 +17,14 @@ scikit-learn>=1.0.2
 scipy>=1.7.3
 
 [:python_version < "3.11"]
-pandas<3.0.0,>=1.4.0
+pandas<3,>=1.4.0
+torch>=1.13.0
+
+[:python_version < "3.14"]
+cloudpickle>=2.1.0
+rdt>=1.18.2
+sdmetrics>=0.21.0
+sdv>=1.21.0
 
 [:python_version >= "3.10" and python_version < "3.11"]
 scikit-learn>=1.1.0
@@ -32,20 +34,34 @@ numpy>=1.24.0
 scipy>=1.9.2
 
 [:python_version >= "3.11" and python_version < "3.12"]
-pandas<3.0.0,>=1.5.0
+pandas<3,>=1.5.0
 scikit-learn>=1.1.3
+torch>=2.0.0
 
 [:python_version >= "3.12" and python_version < "3.13"]
 numpy>=1.26.0
-pandas<3.0.0,>=2.1.1
+pandas<3,>=2.1.1
 scikit-learn>=1.3.1
 scipy>=1.12.0
+torch>=2.3.0
 
-[:python_version >= "3.13"]
+[:python_version >= "3.13" and python_version < "3.14"]
 numpy>=2.1.0
-pandas<3.0.0,>=2.2.3
+pandas<3,>=2.2.3
 scikit-learn>=1.5.2
 scipy>=1.14.1
+torch>=2.6.0
+
+[:python_version >= "3.14"]
+cloudpickle>=3.1.1
+numpy>=2.3.2
+pandas<3,>=2.3.3
+scikit-learn>=1.8.0
+scipy>=1.16.1
+torch>=2.9.0
+rdt>=1.20.0
+sdmetrics>=0.26.0
+sdv>=1.33.0
 
 [all]
 sdgym[dask,dev,test]

{sdgym-0.12.2.dev0 → sdgym-0.13.1.dev0}/tests/test_tasks.py

@@ -118,7 +118,7 @@ def _get_example_pyproject_dict():
                 ],
             },
             'readme': 'README.md',
-            'requires-python': '>=3.9,<3.13',
+            'requires-python': '>=3.9,<3.15',
         },
         'tool': {
             'bumpversion': {

sdgym-0.12.2.dev0/sdgym/run_benchmark/run_benchmark.py

@@ -1,152 +0,0 @@
-"""Script to run a benchmark and upload results to S3."""
-
-import json
-import os
-from datetime import datetime, timezone
-
-from botocore.exceptions import ClientError
-
-from sdgym._benchmark.benchmark import (
-    _benchmark_multi_table_compute_gcp,
-    _benchmark_single_table_compute_gcp,
-)
-from sdgym.run_benchmark.utils import (
-    KEY_DATE_FILE,
-    OUTPUT_DESTINATION_AWS,
-    SYNTHESIZERS_SPLIT_MULTI_TABLE,
-    SYNTHESIZERS_SPLIT_SINGLE_TABLE,
-    _parse_args,
-    get_result_folder_name,
-    post_benchmark_launch_message,
-)
-from sdgym.s3 import get_s3_client, parse_s3_path
-
-MODALITY_TO_SETUP = {
-    'single_table': {
-        'method': _benchmark_single_table_compute_gcp,
-        'synthesizers_split': SYNTHESIZERS_SPLIT_SINGLE_TABLE,
-        'datasets': [
-            'adult',
-            'alarm',
-            'census',
-            'child',
-            'covtype',
-            'expedia_hotel_logs',
-            'insurance',
-            'intrusion',
-            'news',
-        ],
-    },
-    'multi_table': {
-        'method': _benchmark_multi_table_compute_gcp,
-        'synthesizers_split': SYNTHESIZERS_SPLIT_MULTI_TABLE,
-        'datasets': [
-            'WebKP',
-            'DCG',
-            'UW_std',
-            'Same_gen',
-            'CORA',
-            'got_families',
-            'SalesDB',
-            'UTube',
-            'Student_loan',
-            'Hepatitis_std',
-            'Elti',
-            'Bupa',
-            'Toxicology',
-            'imdb_ijs',
-            'ftp',
-            'imdb_small',
-            'imdb_MovieLens',
-            'Pima',
-            'university',
-            'legalActs',
-            'Dunur',
-            'Mesh',
-            'world',
-            'airbnb-simplified',
-            'trains',
-            'FNHK',
-            'fake_hotels',
-            'SAT',
-            'genes',
-            'Biodegradability',
-            'Pyrimidine',
-            'mutagenesis',
-            'restbase',
-            'Triazine',
-            'Carcinogenesis',
-            'fake_hotels_extended',
-            'Mooney_Family',
-            'PTE',
-            'Facebook',
-            'multi_table_ID_demo_dataset',
-            'SAP',
-            'Chess',
-            'Countries',
-            'NCAA',
-            'Atherosclerosis',
-            'nations',
-            'TubePricing',
-            'financial',
-            'Accidents',
-            'MuskSmall',
-            'NBA',
-            'AustralianFootball',
-            'PremierLeague',
-            'OMOP_CDM_dayz',
-        ],
-    },
-}
-
-
-def append_benchmark_run(
-    aws_access_key_id, aws_secret_access_key, date_str, modality='single_table'
-):
-    """Append a new benchmark run to the benchmark dates file in S3."""
-    s3_client = get_s3_client(
-        aws_access_key_id=aws_access_key_id,
-        aws_secret_access_key=aws_secret_access_key,
-    )
-    bucket, prefix = parse_s3_path(OUTPUT_DESTINATION_AWS)
-    try:
-        object = s3_client.get_object(Bucket=bucket, Key=f'{prefix}{modality}/{KEY_DATE_FILE}')
-        body = object['Body'].read().decode('utf-8')
-        data = json.loads(body)
-    except ClientError as e:
-        if e.response['Error']['Code'] == 'NoSuchKey':
-            data = {'runs': []}
-        else:
-            raise RuntimeError(f'Failed to read {KEY_DATE_FILE} from S3: {e}')
-
-    data['runs'].append({'date': date_str, 'folder_name': get_result_folder_name(date_str)})
-    data['runs'] = sorted(data['runs'], key=lambda x: x['date'])
-    s3_client.put_object(
-        Bucket=bucket,
-        Key=f'{prefix}{modality}/{KEY_DATE_FILE}',
-        Body=json.dumps(data).encode('utf-8'),
-    )
-
-
-def main():
-    """Main function to run the benchmark and upload results."""
-    args = _parse_args()
-    aws_access_key_id = os.getenv('AWS_ACCESS_KEY_ID')
-    aws_secret_access_key = os.getenv('AWS_SECRET_ACCESS_KEY')
-    date_str = datetime.now(timezone.utc).strftime('%Y-%m-%d')
-    modality = args.modality
-    for synthesizer_group in MODALITY_TO_SETUP[modality]['synthesizers_split']:
-        MODALITY_TO_SETUP[modality]['method'](
-            output_destination=OUTPUT_DESTINATION_AWS,
-            credential_filepath=os.getenv('CREDENTIALS_FILEPATH'),
-            synthesizers=synthesizer_group,
-            sdv_datasets=MODALITY_TO_SETUP[modality]['datasets'],
-            timeout=345600,  # 4 days
-        )
-
-    append_benchmark_run(aws_access_key_id, aws_secret_access_key, date_str, modality=modality)
-    post_benchmark_launch_message(date_str, compute_service='GCP', modality=modality)
-
-
-if __name__ == '__main__':
-    main()