Train Dim. Reduction — mlTrainDimReduction

Fit PCA as a saved model rather than producing a components table.

What it does

Connect training rows to in. The model output is a one-row reference, not the training data or a live Python estimator. It records mlflow_run_id, model_uri, task_type, model_type, target_column, feature_columns_json and model_config_json. Run metadata contains the model URI, run ID and the training metrics described below.

Runtime fits a scikit-learn pipeline, checks its serialized size and saves it to MLflow. If saving fails, the node fails rather than returning a usable model reference. Preview skips fitting and logging and emits a placeholder with null run ID and URI; configuration and schema checks can still fail during preview.

The built-in model type is pca_fit. Its fitted preprocessing and PCA are saved together. This node returns the model reference; it does not add component columns to the incoming frame. For an immediate components table, use Reduce Dimensions, which fits on its incoming frame without saving a model artifact.

Training metadata contains explained_variance, the sum of the fitted PCA explained-variance ratios. It describes variation retained in the preprocessed training features and is not a predictive-quality score.

PCA exposes transform, not predict; the current Predict and Feature Importance nodes do not consume this artifact. A trusted sklearn loader can load the saved pipeline for use with transform in Python. MLflow may warn that a PCA model lacks a prediction signature or python_function flavor; the sklearn artifact can still be saved and loaded successfully.

Configuration

Omitted or empty features select every input column. Pin the intended features and exclude IDs and known labels yourself. n_components belongs in hyperparameters; choose a value compatible with the fitted feature width and training-row count.

Preprocessing and limits

With omitted or empty preprocessing, numeric features use median imputation and standard scaling. Categorical features use most-frequent imputation and one-hot encoding with unknown categories ignored. These steps are fitted on this node's training rows and travel inside the saved pipeline.

For a nonempty preprocessing object, supply numeric_columns and/or categorical_columns: absent lists are empty and columns outside those lists are dropped. If both lists are empty, preprocessing is omitted and the raw selected features reach the estimator. Numeric options are numeric_strategy (standard_scaler, minmax_scaler, robust_scaler, or another value to omit scaling), impute_numeric (default median) and impute_categorical (default most_frequent).

Runtime requires at least ten rows and warns below fifty. Default host limits are 5,000,000 training rows, 500 selected features and 500 MB of serialized model data. Row and feature limits apply before fitting; model size is checked after fitting and before saving. Configure them with CIAREN_ML_MAX_TRAINING_ROWS, CIAREN_ML_MAX_FEATURE_COLUMNS and CIAREN_ML_MAX_MODEL_SIZE_MB. The default hyperparameter numeric bound is 100,000 (CIAREN_ML_MAX_HYPERPARAMETER_VALUE).

MLflow tracking defaults to ./mlruns; set CIAREN_MLFLOW_TRACKING_URI for a configured store. Load only trusted artifacts: MLflow/joblib deserialization can execute Python code, and URI/path checks do not sandbox that code.

Generated Python code

Download the customer sample CSV and save it as customers.csv beside the script. The sample has 16 rows; Train/Test Split uses test_size=0.25 and seed=42 to keep 12 training rows separate from four held-out rows. Install Ciaren with its ML dependencies to run the scripts. Polars exports cross into pandas/scikit-learn for fitting.

These are complete exports from the flow generator. They fit a pipeline in memory; they do not create an MLflow run, save a model artifact, reproduce run metadata metrics or enforce Ciaren's runtime row/feature/model-size guards. The editor's trained-model reference becomes the actual pipeline variable in the script. To retain an exported model, add your own trusted persistence step.

The example fits two components on tenure, monthly_charges and support_calls, using only 12 training rows. On a parallel branch, File Output writes the four unchanged held-out rows to train-dim-reduction.csv. That CSV is a data passthrough, not PCA coordinates. The fitted pipeline remains in df_3 in the pandas script and _eager_5 in the Polars script.

To apply the fitted model in your own Python workflow, call that pipeline's transform with the same ordered feature columns. Do not call fit_transform on new rows, which fits a new PCA. This sample retains approximately 0.985583 of the preprocessed training variance; the full generator exports below do not append transformations or persistence steps.

The Train configuration is:

{
  "model_type": "pca_fit",
  "feature_columns": [
    "tenure",
    "monthly_charges",
    "support_calls"
  ],
  "hyperparameters": {
    "n_components": 2
  },
  "seed": 42
}

pandas

import pandas as pd
import joblib
from sklearn.compose import ColumnTransformer
from sklearn.decomposition import PCA
from sklearn.impute import SimpleImputer
from sklearn.model_selection import train_test_split
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import OneHotEncoder
from sklearn.preprocessing import StandardScaler

df_customers = pd.read_csv('customers.csv')
df_1, df_2 = train_test_split(df_customers, test_size=0.25, random_state=42)
df_1 = df_1.reset_index(drop=True)
df_2 = df_2.reset_index(drop=True)
_features = ['tenure', 'monthly_charges', 'support_calls']
_X = df_1[_features]
_numeric = [c for c in _features if pd.api.types.is_numeric_dtype(_X[c])]
_categorical = [c for c in _features if c not in _numeric]
_transformers = []
if _numeric:
    _transformers.append(('num', Pipeline([('impute', SimpleImputer(strategy='median')), ('scale', StandardScaler())]), _numeric))
if _categorical:
    _transformers.append(('cat', Pipeline([('impute', SimpleImputer(strategy='most_frequent')), ('encode', OneHotEncoder(handle_unknown='ignore'))]), _categorical))
_preprocessor = ColumnTransformer(_transformers, remainder='drop')
df_3 = Pipeline([('preprocessor', _preprocessor), ('model', PCA(n_components=2, random_state=42))])
df_3.fit(_X)
df_2.to_csv('train-dim-reduction.csv', index=False)

polars

import polars as pl
import joblib
import pandas as pd
from sklearn.compose import ColumnTransformer
from sklearn.decomposition import PCA
from sklearn.impute import SimpleImputer
from sklearn.model_selection import train_test_split
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import OneHotEncoder
from sklearn.preprocessing import StandardScaler

df_customers = pl.read_csv('customers.csv')
_eager_1 = df_customers.to_pandas()
_eager_2, _eager_3 = train_test_split(_eager_1, test_size=0.25, random_state=42)
_eager_2 = _eager_2.reset_index(drop=True)
_eager_3 = _eager_3.reset_index(drop=True)
df_1 = pl.from_pandas(_eager_2)
df_2 = pl.from_pandas(_eager_3)
_eager_4 = df_1.to_pandas()
_features = ['tenure', 'monthly_charges', 'support_calls']
_X = _eager_4[_features]
_numeric = [c for c in _features if pd.api.types.is_numeric_dtype(_X[c])]
_categorical = [c for c in _features if c not in _numeric]
_transformers = []
if _numeric:
    _transformers.append(('num', Pipeline([('impute', SimpleImputer(strategy='median')), ('scale', StandardScaler())]), _numeric))
if _categorical:
    _transformers.append(('cat', Pipeline([('impute', SimpleImputer(strategy='most_frequent')), ('encode', OneHotEncoder(handle_unknown='ignore'))]), _categorical))
_preprocessor = ColumnTransformer(_transformers, remainder='drop')
_eager_5 = Pipeline([('preprocessor', _preprocessor), ('model', PCA(n_components=2, random_state=42))])
_eager_5.fit(_X)
df_2.write_csv('train-dim-reduction.csv')

Tips and common mistakes

• Choose Train Dim. Reduction when you need a reusable saved PCA pipeline; choose Reduce Dimensions when you need transformed rows in the flow.
• PCA does not produce predictions or tree/linear importance. A model wire alone does not make it compatible with Predict or Feature Importance.
• Explain variance in relation to the scaled/encoded features actually fitted, rather than treating it as a percentage of prediction accuracy.
• The sample CSV intentionally contains the held-out input unchanged. Training and transformation are separate operations.

See also

• Reduce Dimensions
• Scale Features
• Select Features
• Train/Test Split
• ML quick start