Merge pull request #13 from soda-inria/benchmark

Benchmark
soda-inria · Dec 17, 2024 · b53a890 · b53a890
2 parents 7662b08 + 9990a01
commit b53a890
Show file tree

Hide file tree

Showing 3 changed files with 6 additions and 3 deletions.
diff --git a/...lts/results_carte_baseline_bestparams.csv → ...lts/results_carte_baseline_bestparams.csv b/...lts/results_carte_baseline_bestparams.csv → ...lts/results_carte_baseline_bestparams.csv
diff --git a/...ts/results_carte_baseline_singletable.csv → ...ts/results_carte_baseline_singletable.csv b/...ts/results_carte_baseline_singletable.csv → ...ts/results_carte_baseline_singletable.csv
diff --git a/carte_ai/scripts/evaluate_singletable.py b/carte_ai/scripts/evaluate_singletable.py
@@ -47,6 +47,7 @@
     RESNETRegressor,
     RESNETClassifier,
 )
+from huggingface_hub import hf_hub_download
 
 
 def _load_data(data_name):
@@ -76,7 +77,7 @@ def _prepare_carte_gnn(
     random_state,
 ):
     """Preprocess for CARTE (graph construction)."""
-    from carte_table_to_graph_old import Table2GraphTransformer
+    from carte_ai.src.carte_table_to_graph import Table2GraphTransformer
 
     data_ = data.copy()
     X_train, X_test, y_train, y_test = set_split(
@@ -85,7 +86,9 @@ def _prepare_carte_gnn(
         num_train,
         random_state=random_state,
     )
-    preprocessor = Table2GraphTransformer()
+    fasttext_path = hf_hub_download(repo_id="hi-paris/fastText", filename="cc.en.300.bin")
+    preprocessor = Table2GraphTransformer(lm_model="fasttext",
+        fasttext_model_path=fasttext_path)
     X_train = preprocessor.fit_transform(X_train, y=y_train)
     X_test = preprocessor.transform(X_test)
     return X_train, X_test, y_train, y_test
@@ -147,7 +150,7 @@ def _prepare_tablevectorizer(
     categorical_preprocessor = Pipeline(
         steps=[
             ("imputer", SimpleImputer(strategy="constant", fill_value="missing")),
-            ("onehot", OneHotEncoder(handle_unknown="ignore", sparse=False)),
+            ("onehot", OneHotEncoder(handle_unknown="ignore", sparse_output=False)),
         ]
     )