drift-ml 0.2.8__tar.gz → 0.2.10__tar.gz

{drift_ml-0.2.8 → drift_ml-0.2.10}/PKG-INFO

@@ -1,13 +1,13 @@
 Metadata-Version: 2.4
 Name: drift-ml
-Version: 0.2.8
+Version: 0.2.10
 Summary: drift — terminal-first, chat-based AutoML. Open source. No tokens. No auth.
+License-Expression: MIT
 Project-URL: Homepage, https://github.com/lakshitsachdeva/intent2model
 Project-URL: Repository, https://github.com/lakshitsachdeva/intent2model
 Keywords: automl,ml,machine-learning,cli,chat,local-first
 Classifier: Development Status :: 4 - Beta
 Classifier: Intended Audience :: Developers
-Classifier: License :: OSI Approved :: MIT License
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
@@ -17,6 +17,9 @@ Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: requests>=2.28.0
+Provides-Extra: dev
+Requires-Dist: pytest>=7.0.0; extra == "dev"
+Requires-Dist: httpx>=0.24.0; extra == "dev"
 Dynamic: license-file
 
 # drift

{drift_ml-0.2.8 → drift_ml-0.2.10}/drift/__init__.py

@@ -16,7 +16,7 @@ Library:
     result = d.train()
 """
 
-__version__ = "0.2.8"
+__version__ = "0.2.10"
 
 from drift.api import Drift
 from drift.cli.client import BackendClient, BackendError

{drift_ml-0.2.8 → drift_ml-0.2.10}/drift/engine_launcher.py

@@ -16,7 +16,7 @@ except ImportError:
     requests = None
 
 GITHUB_REPO = "lakshitsachdeva/intent2model"  # Engine binaries (same repo)
-ENGINE_TAG = "v0.2.8"  # Pinned — direct URL, no API, no rate limits
+ENGINE_TAG = "v0.2.10"  # Pinned — direct URL, no API, no rate limits
 ENGINE_PORT = os.environ.get("DRIFT_ENGINE_PORT", "8000")
 HEALTH_URL = f"http://127.0.0.1:{ENGINE_PORT}/health"
 
@@ -224,4 +224,9 @@ def ensure_engine() -> bool:
     err = stderr_file.read_text().strip() if stderr_file.exists() else ""
     if err:
         print(f"drift: Engine log: {err[-400:]}", file=sys.stderr)
+    if platform.system() == "Windows":
+        print("drift: On Windows, try running the backend manually:", file=sys.stderr)
+        print("  cd backend && python -m uvicorn main:app --host 0.0.0.0 --port 8000", file=sys.stderr)
+        print("  Create .env in project root with GEMINI_API_KEY=... for LLM planning.", file=sys.stderr)
+        print("  Then: set DRIFT_BACKEND_URL=http://localhost:8000", file=sys.stderr)
     return False

{drift_ml-0.2.8 → drift_ml-0.2.10}/drift_ml.egg-info/PKG-INFO

@@ -1,13 +1,13 @@
 Metadata-Version: 2.4
 Name: drift-ml
-Version: 0.2.8
+Version: 0.2.10
 Summary: drift — terminal-first, chat-based AutoML. Open source. No tokens. No auth.
+License-Expression: MIT
 Project-URL: Homepage, https://github.com/lakshitsachdeva/intent2model
 Project-URL: Repository, https://github.com/lakshitsachdeva/intent2model
 Keywords: automl,ml,machine-learning,cli,chat,local-first
 Classifier: Development Status :: 4 - Beta
 Classifier: Intended Audience :: Developers
-Classifier: License :: OSI Approved :: MIT License
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
@@ -17,6 +17,9 @@ Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: requests>=2.28.0
+Provides-Extra: dev
+Requires-Dist: pytest>=7.0.0; extra == "dev"
+Requires-Dist: httpx>=0.24.0; extra == "dev"
 Dynamic: license-file
 
 # drift

{drift_ml-0.2.8 → drift_ml-0.2.10}/drift_ml.egg-info/SOURCES.txt

@@ -18,4 +18,8 @@ drift_ml.egg-info/SOURCES.txt
 drift_ml.egg-info/dependency_links.txt
 drift_ml.egg-info/entry_points.txt
 drift_ml.egg-info/requires.txt
-drift_ml.egg-info/top_level.txt
+drift_ml.egg-info/top_level.txt
+tests/test_api.py
+tests/test_automl_agent.py
+tests/test_evaluator.py
+tests/test_plan_normalizer.py

drift_ml-0.2.10/drift_ml.egg-info/requires.txt

@@ -0,0 +1,5 @@
+requests>=2.28.0
+
+[dev]
+pytest>=7.0.0
+httpx>=0.24.0

{drift_ml-0.2.8 → drift_ml-0.2.10}/pyproject.toml

@@ -4,9 +4,10 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "drift-ml"
-version = "0.2.8"
+version = "0.2.10"
 description = "drift — terminal-first, chat-based AutoML. Open source. No tokens. No auth."
 requires-python = ">=3.10"
+license = "MIT"
 dependencies = [
     "requests>=2.28.0",
 ]
@@ -15,7 +16,6 @@ keywords = ["automl", "ml", "machine-learning", "cli", "chat", "local-first"]
 classifiers = [
     "Development Status :: 4 - Beta",
     "Intended Audience :: Developers",
-    "License :: OSI Approved :: MIT License",
     "Programming Language :: Python :: 3",
     "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
@@ -33,3 +33,10 @@ drift = "drift.__main__:main"
 [tool.setuptools.packages.find]
 where = ["."]
 include = ["drift*"]
+
+[project.optional-dependencies]
+dev = ["pytest>=7.0.0", "httpx>=0.24.0"]
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]
+pythonpath = ["backend"]

drift_ml-0.2.10/tests/test_api.py

@@ -0,0 +1,29 @@
+"""Tests for FastAPI endpoints."""
+import pytest
+from fastapi.testclient import TestClient
+
+
+@pytest.fixture
+def client():
+    """Create test client. Import app after path setup."""
+    import sys
+    from pathlib import Path
+    backend = Path(__file__).resolve().parent.parent / "backend"
+    if str(backend) not in sys.path:
+        sys.path.insert(0, str(backend))
+    from main import app
+    return TestClient(app)
+
+
+def test_health(client):
+    """Health endpoint returns 200."""
+    r = client.get("/health")
+    assert r.status_code == 200
+    data = r.json()
+    assert "status" in data or "llm_available" in data
+
+
+def test_root(client):
+    """Root returns message."""
+    r = client.get("/")
+    assert r.status_code == 200

drift_ml-0.2.10/tests/test_automl_agent.py

@@ -0,0 +1,72 @@
+"""Tests for automl_agent."""
+import pytest
+from agents.automl_agent import _extract_json, _map_alternate_schema
+from agents.plan_normalizer import normalize_plan_dict
+from schemas.pipeline_schema import AutoMLPlan
+
+
+def test_extract_json_from_fenced():
+    """Extract JSON from markdown fenced block."""
+    text = '''```json
+{"inferred_target": "x", "task_type": "regression"}
+```'''
+    out = _extract_json(text)
+    assert out["inferred_target"] == "x"
+    assert out["task_type"] == "regression"
+
+
+def test_extract_json_plain():
+    """Extract JSON from plain text."""
+    text = 'Some text {"inferred_target": "y", "task_type": "classification"} more'
+    out = _extract_json(text)
+    assert out["inferred_target"] == "y"
+    assert out["task_type"] == "classification"
+
+
+def test_extract_json_raises_on_no_json():
+    """Raises when no JSON object found."""
+    with pytest.raises(ValueError, match="No JSON object"):
+        _extract_json("no json here at all")
+
+
+def test_map_alternate_schema_dataset_config():
+    """Map wrong schema (dataset_config) to our schema."""
+    wrong = {
+        "task_description": None,
+        "dataset_config": {"target_column": None, "problem_type": None, "features": []},
+        "preprocessing_pipeline": [],
+        "model_selection_strategy": {"algorithms": []},
+    }
+    profile = {
+        "columns": ["a", "b", "variety"],
+        "dtypes": {"a": "float64", "b": "float64", "variety": "object"},
+        "nunique": {"a": 35, "b": 23, "variety": 3},
+        "numeric_cols": ["a", "b"],
+        "identifier_like_columns": [],
+    }
+    mapped = _map_alternate_schema(wrong, profile, "variety")
+    assert mapped["inferred_target"] == "variety"
+    assert mapped["task_type"] in ("binary_classification", "multiclass_classification")
+    assert len(mapped["model_candidates"]) > 0
+
+
+def test_full_pipeline_alternate_schema_to_automl_plan():
+    """Map alternate schema -> normalize -> AutoMLPlan validates."""
+    wrong = {
+        "dataset_config": {"target_column": None, "problem_type": None},
+        "preprocessing_pipeline": [],
+        "model_selection_strategy": {"algorithms": []},
+    }
+    profile = {
+        "columns": ["a", "b", "y"],
+        "dtypes": {"a": "float64", "b": "float64", "y": "int64"},
+        "nunique": {"a": 10, "b": 8, "y": 2},
+        "numeric_cols": ["a", "b"],
+        "identifier_like_columns": [],
+        "missing_percent": {},
+    }
+    mapped = _map_alternate_schema(wrong, profile, "y")
+    norm = normalize_plan_dict(mapped, profile=profile, requested_target="y")
+    plan = AutoMLPlan(**norm)
+    assert plan.inferred_target == "y"
+    assert plan.task_type == "binary_classification"

drift_ml-0.2.10/tests/test_evaluator.py

@@ -0,0 +1,31 @@
+"""Tests for ml.evaluator."""
+import pytest
+import pandas as pd
+from ml.evaluator import prune_features_aggressive, evaluate_dataset
+
+
+def test_prune_features_drops_constant():
+    """Constant columns are dropped."""
+    df = pd.DataFrame({"a": [1, 2, 3], "b": [5, 5, 5], "target": [0, 1, 0]})
+    out, dropped = prune_features_aggressive(df, "target", "classification")
+    assert "b" in dropped
+    assert "b" not in out.columns
+
+
+def test_prune_features_keeps_target():
+    """Target is never dropped."""
+    df = pd.DataFrame({"x": [1, 2, 3], "target": [0, 1, 0]})
+    out, _ = prune_features_aggressive(df, "target", "classification")
+    assert "target" in out.columns
+
+
+def test_evaluate_dataset_regression():
+    """evaluate_dataset runs for regression."""
+    df = pd.DataFrame({
+        "a": [1.0, 2.0, 3.0, 4.0, 5.0] * 4,
+        "b": [2.0, 4.0, 6.0, 8.0, 10.0] * 4,
+        "target": [3.0, 6.0, 9.0, 12.0, 15.0] * 4,
+    })
+    result = evaluate_dataset(df, "target", "regression")
+    assert "warnings" in result
+    assert isinstance(result["warnings"], list)

drift_ml-0.2.10/tests/test_plan_normalizer.py

@@ -0,0 +1,52 @@
+"""Tests for plan_normalizer."""
+import pytest
+from agents.plan_normalizer import normalize_plan_dict
+
+
+def test_normalize_plan_dict_minimal():
+    """Minimal plan with profile gets filled correctly."""
+    profile = {
+        "columns": ["a", "b", "target"],
+        "dtypes": {"a": "float64", "b": "int64", "target": "object"},
+        "nunique": {"a": 10, "b": 5, "target": 3},
+        "numeric_cols": ["a", "b"],
+        "identifier_like_columns": [],
+        "missing_percent": {},
+    }
+    plan = {
+        "inferred_target": "target",
+        "task_type": "multiclass_classification",
+        "model_candidates": [{"model_name": "logistic_regression", "reason_md": "Baseline.", "params": {}}],
+    }
+    out = normalize_plan_dict(plan, profile=profile)
+    assert out["inferred_target"] == "target"
+    assert out["task_type"] == "multiclass_classification"
+    assert len(out["feature_transforms"]) >= 2  # a, b (target dropped from features)
+    assert len(out["model_candidates"]) > 0
+
+
+def test_normalize_plan_dict_column_name_alias():
+    """column_name is normalized to name."""
+    profile = {"columns": ["x", "y"], "dtypes": {}, "nunique": {}, "numeric_cols": ["x", "y"], "identifier_like_columns": [], "missing_percent": {}}
+    plan = {
+        "inferred_target": "y",
+        "task_type": "regression",
+        "feature_transforms": [{"column_name": "x", "kind": "continuous", "drop": False}],
+    }
+    out = normalize_plan_dict(plan, profile=profile)
+    assert out["feature_transforms"][0]["name"] == "x"
+
+
+def test_normalize_plan_dict_infers_target_from_profile():
+    """When inferred_target missing, uses profile."""
+    profile = {
+        "columns": ["a", "b", "c"],
+        "identifier_like_columns": [],
+        "dtypes": {},
+        "nunique": {"c": 2},
+        "numeric_cols": ["a", "b"],
+        "missing_percent": {},
+    }
+    plan = {"task_type": "binary_classification"}
+    out = normalize_plan_dict(plan, profile=profile)
+    assert out["inferred_target"] == "c"

drift_ml-0.2.8/drift_ml.egg-info/requires.txt

@@ -1 +0,0 @@
-requests>=2.28.0