mostlyai-engine 1.0.0__tar.gz → 1.0.2__tar.gz

{mostlyai_engine-1.0.0 → mostlyai_engine-1.0.2}/PKG-INFO

@@ -1,10 +1,10 @@
 Metadata-Version: 2.4
 Name: mostlyai-engine
-Version: 1.0.0
+Version: 1.0.2
 Summary: Synthetic Data - Engine
-Project-URL: Homepage, https://github.com/mostly-ai/mostlyai-engine
-Project-URL: Documentation, https://mostly-ai.github.io/mostlyai-engine/
-Project-URL: Source, https://github.com/mostly-ai/mostlyai-engine
+Project-URL: homepage, https://github.com/mostly-ai/mostlyai-engine
+Project-URL: repository, https://github.com/mostly-ai/mostlyai-engine
+Project-URL: documentation, https://mostly-ai.github.io/mostlyai-engine/
 Author-email: MOSTLY AI <dev@mostly.ai>
 License-Expression: Apache-2.0
 License-File: LICENSE
@@ -17,9 +17,13 @@ Classifier: Intended Audience :: Science/Research
 Classifier: Intended Audience :: Telecommunications Industry
 Classifier: License :: OSI Approved :: Apache Software License
 Classifier: Operating System :: OS Independent
+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: Topic :: Software Development :: Libraries
 Classifier: Typing :: Typed
-Requires-Python: <3.13,>=3.10
+Requires-Python: >=3.10
 Requires-Dist: accelerate<0.32,>=0.31.0
 Requires-Dist: datasets<3,>=2.20.0
 Requires-Dist: formatron<0.5,>=0.4.11
@@ -29,7 +33,7 @@ Requires-Dist: numpy<2,>=1.26.3
 Requires-Dist: opacus<2,>=1.5.2
 Requires-Dist: pandas~=2.2.0
 Requires-Dist: peft<0.12,>=0.11.1
-Requires-Dist: psutil<6,>=5.9.8
+Requires-Dist: psutil<6,>=5.9.5
 Requires-Dist: pyarrow>=16.0.0
 Requires-Dist: rich<14,>=13.9.4
 Requires-Dist: setuptools<76,>=75.0.0
@@ -77,7 +81,7 @@ pip install -U mostlyai-engine
 
 or alternatively for a GPU setup:
 ```bash
-pip install -U mostlyai-engine[gpu]
+pip install -U 'mostlyai-engine[gpu]'
 ```
 
 
@@ -90,8 +94,9 @@ from pathlib import Path
 import pandas as pd
 from mostlyai import engine
 
-# set up workspace
+# set up workspace and default logging
 ws = Path("ws-tabular-flat")
+engine.init_logging()
 
 # load original data
 url = "https://github.com/mostly-ai/public-demo-data/raw/refs/heads/dev/census"
@@ -105,7 +110,10 @@ engine.split(                         # split data as PQT files for `trn` + `val
 )
 engine.analyze(workspace_dir=ws)      # generate column-level statistics to `{ws}/ModelData/tgt-stats/stats.json`
 engine.encode(workspace_dir=ws)       # encode training data to `{ws}/OriginalData/encoded-data`
-engine.train(workspace_dir=ws)        # train model and store to `{ws}/ModelData/model-data`
+engine.train(                         # train model and store to `{ws}/ModelStore/model-data`
+    workspace_dir=ws,
+    max_training_time=1,              # limit TRAIN to 1 minute for demo purposes
+)
 engine.generate(workspace_dir=ws)     # use model to generate synthetic samples to `{ws}/SyntheticData`
 pd.read_parquet(ws / "SyntheticData") # load synthetic data
 ```
@@ -117,8 +125,11 @@ from pathlib import Path
 import pandas as pd
 from mostlyai import engine
 
-# set up workspace
+engine.init_logging()
+
+# set up workspace and default logging
 ws = Path("ws-tabular-sequential")
+engine.init_logging()
 
 # load original data
 url = "https://github.com/mostly-ai/public-demo-data/raw/refs/heads/dev/baseball"
@@ -138,7 +149,7 @@ engine.analyze(workspace_dir=ws)      # generate column-level statistics to `{ws
 engine.encode(workspace_dir=ws)       # encode training data to `{ws}/OriginalData/encoded-data`
 engine.train(                         # train model and store to `{ws}/ModelStore/model-data`
     workspace_dir=ws,
-    max_training_time=2,              # limit TRAIN to 1 minute for demo purposes
+    max_training_time=1,              # limit TRAIN to 1 minute for demo purposes
 )
 engine.generate(workspace_dir=ws)     # use model to generate synthetic samples to `{ws}/SyntheticData`
 pd.read_parquet(ws / "SyntheticData") # load synthetic data
@@ -151,8 +162,9 @@ from pathlib import Path
 import pandas as pd
 from mostlyai import engine
 
-# set up workspace
+# init workspace and logging
 ws = Path("ws-language-flat")
+engine.init_logging()
 
 # load original data
 trn_df = pd.read_parquet("https://github.com/mostly-ai/public-demo-data/raw/refs/heads/dev/headlines/headlines.parquet")

{mostlyai_engine-1.0.0 → mostlyai_engine-1.0.2}/README.md

@@ -30,7 +30,7 @@ pip install -U mostlyai-engine
 
 or alternatively for a GPU setup:
 ```bash
-pip install -U mostlyai-engine[gpu]
+pip install -U 'mostlyai-engine[gpu]'
 ```
 
 
@@ -43,8 +43,9 @@ from pathlib import Path
 import pandas as pd
 from mostlyai import engine
 
-# set up workspace
+# set up workspace and default logging
 ws = Path("ws-tabular-flat")
+engine.init_logging()
 
 # load original data
 url = "https://github.com/mostly-ai/public-demo-data/raw/refs/heads/dev/census"
@@ -58,7 +59,10 @@ engine.split(                         # split data as PQT files for `trn` + `val
 )
 engine.analyze(workspace_dir=ws)      # generate column-level statistics to `{ws}/ModelData/tgt-stats/stats.json`
 engine.encode(workspace_dir=ws)       # encode training data to `{ws}/OriginalData/encoded-data`
-engine.train(workspace_dir=ws)        # train model and store to `{ws}/ModelData/model-data`
+engine.train(                         # train model and store to `{ws}/ModelStore/model-data`
+    workspace_dir=ws,
+    max_training_time=1,              # limit TRAIN to 1 minute for demo purposes
+)
 engine.generate(workspace_dir=ws)     # use model to generate synthetic samples to `{ws}/SyntheticData`
 pd.read_parquet(ws / "SyntheticData") # load synthetic data
 ```
@@ -70,8 +74,11 @@ from pathlib import Path
 import pandas as pd
 from mostlyai import engine
 
-# set up workspace
+engine.init_logging()
+
+# set up workspace and default logging
 ws = Path("ws-tabular-sequential")
+engine.init_logging()
 
 # load original data
 url = "https://github.com/mostly-ai/public-demo-data/raw/refs/heads/dev/baseball"
@@ -91,7 +98,7 @@ engine.analyze(workspace_dir=ws)      # generate column-level statistics to `{ws
 engine.encode(workspace_dir=ws)       # encode training data to `{ws}/OriginalData/encoded-data`
 engine.train(                         # train model and store to `{ws}/ModelStore/model-data`
     workspace_dir=ws,
-    max_training_time=2,              # limit TRAIN to 1 minute for demo purposes
+    max_training_time=1,              # limit TRAIN to 1 minute for demo purposes
 )
 engine.generate(workspace_dir=ws)     # use model to generate synthetic samples to `{ws}/SyntheticData`
 pd.read_parquet(ws / "SyntheticData") # load synthetic data
@@ -104,8 +111,9 @@ from pathlib import Path
 import pandas as pd
 from mostlyai import engine
 
-# set up workspace
+# init workspace and logging
 ws = Path("ws-language-flat")
+engine.init_logging()
 
 # load original data
 trn_df = pd.read_parquet("https://github.com/mostly-ai/public-demo-data/raw/refs/heads/dev/headlines/headlines.parquet")

{mostlyai_engine-1.0.0 → mostlyai_engine-1.0.2}/mostlyai/engine/__init__.py

@@ -13,6 +13,7 @@
 # limitations under the License.
 import warnings
 
+from mostlyai.engine.logging import init_logging
 from mostlyai.engine.splitting import split
 from mostlyai.engine.analysis import analyze
 from mostlyai.engine.training import train
@@ -20,8 +21,8 @@ from mostlyai.engine.encoding import encode
 from mostlyai.engine.generation import generate
 
 
-__all__ = ["split", "analyze", "encode", "train", "generate"]
-__version__ = "1.0.0"
+__all__ = ["split", "analyze", "encode", "train", "generate", "init_logging"]
+__version__ = "1.0.2"
 
 # suppress specific warning related to os.fork() in multi-threaded processes
 warnings.filterwarnings("ignore", category=DeprecationWarning, message=".*multi-threaded.*fork.*")

{mostlyai_engine-1.0.0 → mostlyai_engine-1.0.2}/mostlyai/engine/_language/training.py

@@ -197,8 +197,8 @@ def _calculate_val_loss(model: PreTrainedModel | GradSampleModule, val_dataloade
 def train(
     *,
     model: str = "MOSTLY_AI/LSTMFromScratch-3m",
-    max_training_time: float = 60.0,
-    max_epochs: float = 100.0,
+    max_training_time: float = 14400.0,  # 10 days
+    max_epochs: float = 100.0,  # 100 epochs
     batch_size: int | None = None,
     gradient_accumulation_steps: int | None = None,
     enable_flexible_generation: bool = True,

{mostlyai_engine-1.0.0 → mostlyai_engine-1.0.2}/mostlyai/engine/_tabular/training.py

@@ -339,8 +339,8 @@ def _calculate_average_trn_loss(trn_sample_losses: list[torch.Tensor], n: int |
 def train(
     *,
     model: str = "MOSTLY_AI/Medium",
-    max_training_time: float = 60.0,
-    max_epochs: float = 100.0,
+    max_training_time: float = 14400.0,  # 10 days
+    max_epochs: float = 100.0,  # 100 epochs
     batch_size: int | None = None,
     gradient_accumulation_steps: int | None = None,
     max_sequence_window: int = 100,

{mostlyai_engine-1.0.0 → mostlyai_engine-1.0.2}/mostlyai/engine/_workspace.py

@@ -168,21 +168,6 @@ class Workspace:
         self.generated_data_path = Path(self._ws_path) / Path(*generated_data)
         self.generated_data = make_path_desc(parts=generated_data, fetch_multiple=fetch_part_parquets)
 
-        # Report-related
-        self.report_ctx_data_path = self._ws_path / "report-ctx-data"
-        self.report_ctx_data = make_path_desc(
-            parts=["report-ctx-data"],
-            fetch_multiple=fetch_part_parquets,
-        )
-        self.report_trn_ctx_data = make_path_desc(
-            parts=["report-ctx-data"],
-            fetch_multiple=fetch_part_trn_parquets,
-        )
-        self.report_val_ctx_data = make_path_desc(
-            parts=["report-ctx-data"],
-            fetch_multiple=fetch_part_val_parquets,
-        )
-
 
 def ensure_workspace_dir(workspace_dir: str | Path) -> Path:
     workspace_dir = Path(workspace_dir)

mostlyai_engine-1.0.2/mostlyai/engine/logging.py

@@ -0,0 +1,34 @@
+# Copyright 2025 MOSTLY AI
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import sys
+import logging
+
+_LOG = logging.getLogger(__name__.rsplit(".", 1)[0])  # get the logger with the root module name (mostlyai.engine)
+
+
+def init_logging() -> None:
+    """
+    Initialize the logging configuration. Either log to stdout or to a file.
+    """
+
+    # log to stdout
+    handler = logging.StreamHandler(stream=sys.stdout)
+    handler.setFormatter(logging.Formatter("[%(asctime)s] %(levelname)-7s: %(message)s"))
+    handler.setLevel(logging.INFO)
+
+    if not _LOG.hasHandlers():
+        _LOG.addHandler(handler)
+        _LOG.setLevel(logging.INFO)
+        _LOG.propagate = False

{mostlyai_engine-1.0.0 → mostlyai_engine-1.0.2}/mostlyai/engine/training.py

@@ -26,8 +26,8 @@ from mostlyai.engine._workspace import resolve_model_type
 def train(
     *,
     model: str | None = None,
-    max_training_time: float = 60.0,
-    max_epochs: float = 100.0,
+    max_training_time: float = 14400.0,  # 10 days
+    max_epochs: float = 100.0,  # 100 epochs
     batch_size: int | None = None,
     gradient_accumulation_steps: int | None = None,
     enable_flexible_generation: bool = True,

{mostlyai_engine-1.0.0 → mostlyai_engine-1.0.2}/pyproject.toml

@@ -1,9 +1,9 @@
 [project]
 name = "mostlyai-engine"
-version = "1.0.0"
+version = "1.0.2"
 description = "Synthetic Data - Engine"
 authors = [{ name = "MOSTLY AI", email = "dev@mostly.ai" }]
-requires-python = ">=3.10,<3.13"
+requires-python = ">=3.10"
 readme = "README.md"
 license = "Apache-2.0"
 classifiers = [
@@ -14,6 +14,10 @@ classifiers = [
     "Intended Audience :: Financial and Insurance Industry",
     "Intended Audience :: Healthcare Industry",
     "Intended Audience :: Telecommunications Industry",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
     "License :: OSI Approved :: Apache Software License",
     "Operating System :: OS Independent",
     "Topic :: Software Development :: Libraries",
@@ -27,7 +31,7 @@ dependencies = [
     "torch==2.5.1; sys_platform == 'darwin'",
     "pyarrow>=16.0.0",
     "joblib>=1.3.0,<2",
-    "psutil>=5.9.8,<6",
+    "psutil>=5.9.5,<6",  # upgrade when colab psutil is updated
     "tokenizers>=0.20.1,<0.21",
     "transformers>=4.45.2,<5",
     "datasets>=2.20.0,<3",
@@ -69,9 +73,9 @@ docs = [
 ]
 
 [project.urls]
-Homepage = "https://github.com/mostly-ai/mostlyai-engine"
-Documentation = "https://mostly-ai.github.io/mostlyai-engine/"
-Source = "https://github.com/mostly-ai/mostlyai-engine"
+homepage = "https://github.com/mostly-ai/mostlyai-engine"
+repository = "https://github.com/mostly-ai/mostlyai-engine"
+documentation = "https://mostly-ai.github.io/mostlyai-engine/"
 
 [tool.uv]
 default-groups = ["test", "dev", "docs"]