remove mlem

README.md

@@ -16,7 +16,7 @@ A package to help create and deploy Translator Reasoner APIs (TRAPI) from any pr
 The **TRAPI Predict Kit** helps data scientists to build, and **publish prediction models** in a [FAIR](https://www.go-fair.org/fair-principles/) and reproducible manner. It provides helpers for various steps of the process:
 
 * A template to help user quickly bootstrap a new prediction project with the recommended structure ([MaastrichtU-IDS/cookiecutter-trapi-predict-kit](https://github.com/MaastrichtU-IDS/cookiecutter-trapi-predict-kit/))
-* Helper function to easily save a generated model, its metadata, and the data used to generate it. It uses tools such as [`dvc`](https://dvc.org/) and [`mlem`](https://mlem.ai/) to store large model outside of the git repository.
+* Helper function to easily save a generated model, its metadata, and the data used to generate it. It uses tools such as [`dvc`](https://dvc.org/) to store large model outside of the git repository.
 * Deploy API endpoints for retrieving predictions, which comply with the NCATS Biomedical Data Translator standards ([Translator Reasoner API](https://github.com/NCATSTranslator/ReasonerAPI) and [BioLink model](https://github.com/biolink/biolink-model)), using a decorator `@trapi_predict` to simply annotate the function that produces predicted associations for a given input entity
 
 Predictions are usually generated from machine learning models (e.g. predict disease treated by drug), but it can adapt to generic python function, as long as the input params and return object follow the expected structure.
@@ -182,7 +182,7 @@ uvicorn trapi.main:app --port 8808 --reload
 
 ### 💾 Save a generated model
 
-Helper function to easily save a generated model, its metadata, and the data used to generate it. It uses tools such as [`dvc`](https://dvc.org/) and [`mlem`](https://mlem.ai/) to store large model outside of the git repository.
+Helper function to easily save a generated model, its metadata, and the data used to generate it. It uses tools such as [`dvc`](https://dvc.org/) to store large model outside of the git repository.
 
 ```python
 from trapi_predict_kit import save

pyproject.toml

@@ -35,7 +35,7 @@ dependencies = [
     "fastapi >=0.68.1",
     "rdflib >=6.1.1",
     "reasoner-pydantic >=3.0.1",
-    "mlem",
+    # "mlem",
     "dvc",
     "bmt",
     # "fairworkflows @ git+https://github.com/vemonet/fairworkflows.git",

src/trapi_predict_kit/save.py

@@ -3,7 +3,7 @@
 from dataclasses import dataclass
 from typing import Any, Optional
 
-from mlem import api as mlem
+# from mlem import api as mlem
 from rdflib import Graph
 
 # from mlem.api import save as mlem_save, load as mlem_load
@@ -42,11 +42,11 @@ def save(
 
     # mlem_model = MlemModel.from_obj(model, sample_data=sample_data)
     # mlem_model.dump(path)
-    if method == "mlem":
-        mlem.save(model, path, sample_data=sample_data)
-    else:
-        with open(path, "wb") as f:
-            pickle.dump(model, f)
+    # if method == "mlem":
+    #     mlem.save(model, path, sample_data=sample_data)
+    # else:
+    with open(path, "wb") as f:
+        pickle.dump(model, f)
 
     g = get_run_metadata(scores, sample_data, hyper_params, model_name)
     g.serialize(f"{path}.ttl", format="ttl")
@@ -63,11 +63,11 @@ def save(
 
 def load(path: str, method: str = "pickle") -> LoadedModel:
     log.info(f"💽 Loading model from {path} using {method}")
-    if method == "mlem":
-        model = mlem.load(path)
-    else:
-        with open(path, "rb") as f:
-            model = pickle.load(f)
+    # if method == "mlem":
+    #     model = mlem.load(path)
+    # else:
+    with open(path, "rb") as f:
+        model = pickle.load(f)
 
     g = Graph()
     g.parse(f"{path}.ttl", format="ttl")

tests/test_save.py

@@ -41,12 +41,12 @@ def test_save_pickle():
     shutil.rmtree(tmp_path)
 
 
-def test_save_mlem():
-    """Test to save and load a basic model with mlem"""
-    save(model, model_path, sample_data=data, scores=scores, hyper_params=hyper_params, method="mlem")
-    assert Path(model_path).is_file()
-    assert Path(f"{model_path}.mlem").is_file()
-    assert Path(f"{model_path}.ttl").is_file()
-    loaded_model = load(model_path, method="mlem")
-    assert loaded_model.model is not None
-    shutil.rmtree(tmp_path)
+# def test_save_mlem():
+#     """Test to save and load a basic model with mlem"""
+#     save(model, model_path, sample_data=data, scores=scores, hyper_params=hyper_params, method="mlem")
+#     assert Path(model_path).is_file()
+#     assert Path(f"{model_path}.mlem").is_file()
+#     assert Path(f"{model_path}.ttl").is_file()
+#     loaded_model = load(model_path, method="mlem")
+#     assert loaded_model.model is not None
+#     shutil.rmtree(tmp_path)