mostlyai-engine 1.1.0__tar.gz → 1.1.2__tar.gz

{mostlyai_engine-1.1.0 → mostlyai_engine-1.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mostlyai-engine
-Version: 1.1.0
+Version: 1.1.2
 Summary: Synthetic Data Engine
 Project-URL: homepage, https://github.com/mostly-ai/mostlyai-engine
 Project-URL: repository, https://github.com/mostly-ai/mostlyai-engine
@@ -30,7 +30,7 @@ Requires-Dist: formatron<0.5,>=0.4.11
 Requires-Dist: joblib<2,>=1.3.0
 Requires-Dist: json-repair<0.31,>=0.30.0
 Requires-Dist: numpy<2,>=1.26.3
-Requires-Dist: opacus<2,>=1.5.2
+Requires-Dist: opacus==1.5.2
 Requires-Dist: pandas~=2.2.0
 Requires-Dist: peft<0.12,>=0.11.1
 Requires-Dist: psutil<6,>=5.9.5
@@ -168,13 +168,17 @@ 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")
-trn_df = trn_df.sample(n=10_000, random_state=42)[['category', 'headline']]
+trn_df = trn_df.sample(n=10_000, random_state=42)
 
 # execute the engine steps
 engine.split(                         # split data as PQT files for `trn` + `val` to `{ws}/OriginalData/tgt-data`
     workspace_dir=ws,
     tgt_data=trn_df,
-    model_type="LANGUAGE",
+    tgt_encoding_types={
+        'category': 'LANGUAGE_CATEGORICAL',
+        'date': 'LANGUAGE_DATETIME',
+        'headline': 'LANGUAGE_TEXT',
+    }
 )
 engine.analyze(workspace_dir=ws)      # generate column-level statistics to `{ws}/ModelStore/tgt-stats/stats.json`
 engine.encode(workspace_dir=ws)       # encode training data to `{ws}/OriginalData/encoded-data`

{mostlyai_engine-1.1.0 → mostlyai_engine-1.1.2}/README.md

@@ -117,13 +117,17 @@ 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")
-trn_df = trn_df.sample(n=10_000, random_state=42)[['category', 'headline']]
+trn_df = trn_df.sample(n=10_000, random_state=42)
 
 # execute the engine steps
 engine.split(                         # split data as PQT files for `trn` + `val` to `{ws}/OriginalData/tgt-data`
     workspace_dir=ws,
     tgt_data=trn_df,
-    model_type="LANGUAGE",
+    tgt_encoding_types={
+        'category': 'LANGUAGE_CATEGORICAL',
+        'date': 'LANGUAGE_DATETIME',
+        'headline': 'LANGUAGE_TEXT',
+    }
 )
 engine.analyze(workspace_dir=ws)      # generate column-level statistics to `{ws}/ModelStore/tgt-stats/stats.json`
 engine.encode(workspace_dir=ws)       # encode training data to `{ws}/OriginalData/encoded-data`

{mostlyai_engine-1.1.0 → mostlyai_engine-1.1.2}/mostlyai/engine/__init__.py

@@ -22,7 +22,7 @@ from mostlyai.engine.generation import generate
 
 
 __all__ = ["split", "analyze", "encode", "train", "generate", "init_logging"]
-__version__ = "1.1.0"
+__version__ = "1.1.2"
 
 # suppress specific warning related to os.fork() in multi-threaded processes
 warnings.filterwarnings("ignore", category=DeprecationWarning, message=".*multi-threaded.*fork.*")

{mostlyai_engine-1.1.0 → mostlyai_engine-1.1.2}/mostlyai/engine/splitting.py

@@ -55,8 +55,13 @@ def _get_default_tabular_encoding_type(x: pd.Series) -> ModelEncodingType:
         return ModelEncodingType.tabular_categorical
 
 
-def _get_default_language_encoding_type(_: pd.Series) -> ModelEncodingType:
-    return ModelEncodingType.language_text
+def _get_default_language_encoding_type(x: pd.Series) -> ModelEncodingType:
+    if is_integer_dtype(x) or is_float_dtype(x):
+        return ModelEncodingType.language_numeric
+    elif is_date_dtype(x) or is_timestamp_dtype(x):
+        return ModelEncodingType.language_datetime
+    else:
+        return ModelEncodingType.language_categorical
 
 
 def split(

{mostlyai_engine-1.1.0 → mostlyai_engine-1.1.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "mostlyai-engine"
-version = "1.1.0"
+version = "1.1.2"
 description = "Synthetic Data Engine"
 authors = [{ name = "MOSTLY AI", email = "dev@mostly.ai" }]
 requires-python = ">=3.10"
@@ -36,7 +36,7 @@ dependencies = [
     "datasets>=2.20.0,<3",
     "accelerate>=0.31.0,<0.32",
     "peft>=0.11.1,<0.12",
-    "opacus>=1.5.2,<2",
+    "opacus==1.5.2",  # pin as 1.5.3 breaks tests
     "formatron>=0.4.11,<0.5",
     "json-repair>=0.30.0,<0.31",
     "rich>=13.9.4,<14",