Propagate failures in pandas integration tests and Skip failing tests (…

…#17521)

This PR ensures that the integration tests fail in any one of the test modules fails. It also skips of xfails any tests that are not currently passing. Finally, it fixes one incorrect use of `rng.random`.

Some of the change were originally made in  #17489

Authors:
  - Matthew Murray (https://github.com/Matt711)

Approvers:
  - Vyas Ramasubramani (https://github.com/vyasr)

URL: #17521

ci/cudf_pandas_scripts/third-party-integration/test.sh

@@ -26,6 +26,8 @@ main() {
     LIBS=${LIBS#[}
     LIBS=${LIBS%]}
 
+    ANY_FAILURES=0
+
     for lib in ${LIBS//,/ }; do
         lib=$(echo "$lib" | tr -d '""')
         echo "Running tests for library $lib"
@@ -56,10 +58,6 @@ main() {
         rapids-logger "Check GPU usage"
         nvidia-smi
 
-        EXITCODE=0
-        trap "EXITCODE=1" ERR
-        set +e
-
         rapids-logger "pytest ${lib}"
 
         NUM_PROCESSES=8
@@ -72,12 +70,20 @@ main() {
             fi
         done
 
+        EXITCODE=0
+        trap "EXITCODE=1" ERR
+        set +e
+
         TEST_DIR=${TEST_DIR} NUM_PROCESSES=${NUM_PROCESSES} ci/cudf_pandas_scripts/third-party-integration/run-library-tests.sh ${lib}
 
+        set -e
         rapids-logger "Test script exiting with value: ${EXITCODE}"
+        if [[ ${EXITCODE} != 0 ]]; then
+            ANY_FAILURES=1
+        fi
     done
 
-    exit ${EXITCODE}
+    exit ${ANY_FAILURES}
 }
 
 main "$@"

python/cudf/cudf_pandas_tests/third_party_integration_tests/dependencies.yaml

@@ -76,13 +76,6 @@ files:
       - py_version
       - test_base
       - test_xgboost
-  test_catboost:
-    output: none
-    includes:
-      - cuda_version
-      - py_version
-      - test_base
-      - test_catboost
   test_cuml:
     output: none
     includes:
@@ -251,14 +244,6 @@ dependencies:
           - pip
           - pip:
             - xgboost>=2.0.1
-  test_catboost:
-    common:
-      - output_types: conda
-        packages:
-          - numpy
-          - scipy
-          - scikit-learn
-          - catboost
   test_cuml:
     common:
       - output_types: conda

python/cudf/cudf_pandas_tests/third_party_integration_tests/tests/test_holoviews.py

@@ -71,6 +71,9 @@ def test_holoviews_heatmap(df):
     )
 
 
+@pytest.mark.skip(
+    reason="AttributeError: 'ndarray' object has no attribute '_fsproxy_wrapped'"
+)
 def test_holoviews_histogram(df):
     return get_plot_info(hv.Histogram(df.values))
 

python/cudf/cudf_pandas_tests/third_party_integration_tests/tests/test_matplotlib.py

@@ -33,13 +33,19 @@ def assert_plots_equal(expect, got):
 pytestmark = pytest.mark.assert_eq(fn=assert_plots_equal)
 
 
+@pytest.mark.skip(
+    reason="AttributeError: 'ndarray' object has no attribute '_fsproxy_wrapped'"
+)
 def test_line():
     df = pd.DataFrame({"x": [1, 2, 3, 4, 5], "y": [2, 4, 6, 8, 10]})
     (data,) = plt.plot(df["x"], df["y"], marker="o", linestyle="-")
 
     return plt.gca()
 
 
+@pytest.mark.skip(
+    reason="AttributeError: 'ndarray' object has no attribute '_fsproxy_wrapped'"
+)
 def test_bar():
     data = pd.Series([1, 2, 3, 4, 5], index=["a", "b", "c", "d", "e"])
     ax = data.plot(kind="bar")

python/cudf/cudf_pandas_tests/third_party_integration_tests/tests/test_numpy.py

@@ -37,6 +37,9 @@ def test_numpy_dot(df):
     return np.dot(df, df.T)
 
 
+@pytest.mark.skip(
+    reason="AttributeError: 'ndarray' object has no attribute '_fsproxy_wrapped'"
+)
 def test_numpy_fft(sr):
     fft = np.fft.fft(sr)
     return fft

python/cudf/cudf_pandas_tests/third_party_integration_tests/tests/test_pytorch.py

@@ -116,6 +116,9 @@ def test_torch_train(data):
     return model(test_x1, test_x2)
 
 
+@pytest.mark.skip(
+    reason="AssertionError: The values for attribute 'device' do not match: cpu != cuda:0."
+)
 def test_torch_tensor_ctor():
     s = pd.Series(range(5))
     return torch.tensor(s.values)

python/cudf/cudf_pandas_tests/third_party_integration_tests/tests/test_seaborn.py

@@ -54,6 +54,9 @@ def test_scatter(df):
     return ax
 
 
+@pytest.mark.skip(
+    reason="AttributeError: 'ndarray' object has no attribute '_fsproxy_wrapped'"
+)
 def test_lineplot_with_sns_data():
     df = sns.load_dataset("flights")
     ax = sns.lineplot(data=df, x="month", y="passengers")

python/cudf/cudf_pandas_tests/third_party_integration_tests/tests/test_stumpy_distributed.py

@@ -41,7 +41,7 @@ def test_multidimensional_distributed_timeseries(dask_client):
     rng = np.random.default_rng(seed=42)
     # Each row represents data from a different dimension while each column represents
     # data from the same dimension
-    your_time_series = rng.random(3, 1000)
+    your_time_series = rng.random((3, 1000))
     # Approximately, how many data points might be found in a pattern
     window_size = 50
 

python/cudf/cudf_pandas_tests/third_party_integration_tests/tests/test_tensorflow.py

@@ -271,6 +271,7 @@ def call(self, values):
         return tf.concat(values, axis=-1)
 
 
+@pytest.mark.xfail(reason="ValueError: Invalid dtype: object")
 def test_full_example_train_with_df(df, target):
     # https://www.tensorflow.org/tutorials/load_data/pandas_dataframe#full_example
     # Inputs are directly passed as dictionary of series

python/cudf/cudf_pandas_tests/third_party_integration_tests/tests/test_xgboost.py

@@ -113,6 +113,9 @@ def test_with_external_memory(
     return predt
 
 
+@pytest.mark.skip(
+    reason="TypeError: Implicit conversion to a NumPy array is not allowed. Please use `.get()` to construct a NumPy array explicitly."
+)
 @pytest.mark.parametrize("device", ["cpu", "cuda"])
 def test_predict(device: str) -> np.ndarray:
     reg = xgb.XGBRegressor(n_estimators=2, device=device)