Remove drop_duplicates() from SAR method fix #1464 #1588

miguelgfierro · 2021-12-17T09:24:00Z

Description

Related Issues

Checklist:

I have followed the contribution guidelines and code style for this project.
I have added tests covering my contributions.
I have updated the documentation accordingly.
This PR is being made to staging branch and not to main branch.

miguelgfierro · 2021-12-17T11:39:41Z

error:

tests/unit/recommenders/models/test_sar_singlenode.py ......FFFFFF...... [ 74%]
________________ test_sar_item_similarity[1-cooccurrence-count] ________________

threshold = 1, similarity_type = 'cooccurrence', file = 'count'
demo_usage_data =                  UserId    MovieId     Timestamp  Rating
0      0003000098E85347  DQF-00358  1.433879e+09     1.0
1   ...BAE  DQF-00248  1.416292e+09     1.0
11837  00030000822E3BAE  DAF-00448  1.416292e+09     1.0

[11838 rows x 4 columns]
sar_settings = {'ATOL': 1e-08, 'FILE_DIR': 'https://recodatasets.z20.web.core.windows.net/sarunittest/', 'TEST_USER_ID': '0003000098E85347'}
header = {'col_item': 'MovieId', 'col_rating': 'Rating', 'col_timestamp': 'Timestamp', 'col_user': 'UserId'}

    @pytest.mark.parametrize(
        "threshold,similarity_type,file",
        [
            (1, "cooccurrence", "count"),
            (1, "jaccard", "jac"),
            (1, "lift", "lift"),
            (3, "cooccurrence", "count"),
            (3, "jaccard", "jac"),
            (3, "lift", "lift"),
        ],
    )
    def test_sar_item_similarity(
        threshold, similarity_type, file, demo_usage_data, sar_settings, header
    ):
    
        model = SARSingleNode(
            similarity_type=similarity_type,
            timedecay_formula=False,
            time_decay_coefficient=30,
            threshold=threshold,
            **header
        )
    
        model.fit(demo_usage_data)
    
        true_item_similarity, row_ids, col_ids = read_matrix(
            sar_settings["FILE_DIR"] + "sim_" + file + str(threshold) + ".csv"
        )
    
        if similarity_type == "cooccurrence":
            test_item_similarity = _rearrange_to_test(
                model.item_similarity.todense(),
                row_ids,
                col_ids,
                model.item2index,
                model.item2index,
            )
>           assert np.array_equal(
                true_item_similarity.astype(test_item_similarity.dtype),
                test_item_similarity,
            )
E           AssertionError: assert False
E            +  where False = <function array_equal at 0x7f5b7d41dd40>(array([[ 31.,   0.,   0., ...,   0.,  16.,   0.],\n       [  0.,  29.,   5., ...,   0.,   0.,   0.],\n       [  0.,   5.... 60.,  12.,   1.],\n       [ 16.,   0.,   0., ...,  12., 149.,   0.],\n       [  0.,   0.,   0., ...,   1.,   0.,  29.]]), matrix([[ 55.,   0.,   0., ...,   0.,  20.,   0.],\n        [  0.,  35.,   6., ...,   0.,   0.,   0.],\n        [  0.,  ...3.,  15.,   1.],\n        [ 20.,   0.,   0., ...,  15., 243.,   0.],\n        [  0.,   0.,   0., ...,   1.,   0., 103.]]))
E            +    where <function array_equal at 0x7f5b7d41dd40> = np.array_equal
E            +    and   array([[ 31.,   0.,   0., ...,   0.,  16.,   0.],\n       [  0.,  29.,   5., ...,   0.,   0.,   0.],\n       [  0.,   5.... 60.,  12.,   1.],\n       [ 16.,   0.,   0., ...,  12., 149.,   0.],\n       [  0.,   0.,   0., ...,   1.,   0.,  29.]]) = <built-in method astype of numpy.ndarray object at 0x7f5b407a5b10>(dtype('float64'))
E            +      where <built-in method astype of numpy.ndarray object at 0x7f5b407a5b10> = array([['31', '0', '0', ..., '0', '16', '0'],\n       ['0', '29', '5', ..., '0', '0', '0'],\n       ['0', '5', '55', ...... '12', '1'],\n       ['16', '0', '0', ..., '12', '149', '0'],\n       ['0', '0', '0', ..., '1', '0', '29']], dtype='<U4').astype
E            +      and   dtype('float64') = matrix([[ 55.,   0.,   0., ...,   0.,  20.,   0.],\n        [  0.,  35.,   6., ...,   0.,   0.,   0.],\n        [  0.,  ...3.,  15.,   1.],\n        [ 20.,   0.,   0., ...,  15., 243.,   0.],\n        [  0.,   0.,   0., ...,   1.,   0., 103.]]).dtype

tests/unit/recommenders/models/test_sar_singlenode.py:175: AssertionError
___________________ test_sar_item_similarity[1-jaccard-jac] ____________________

threshold = 1, similarity_type = 'jaccard', file = 'jac'
demo_usage_data =                  UserId    MovieId     Timestamp  Rating
0      0003000098E85347  DQF-00358  1.433879e+09     1.0
1   ...BAE  DQF-00248  1.416292e+09     1.0
11837  00030000822E3BAE  DAF-00448  1.416292e+09     1.0

[11838 rows x 4 columns]
sar_settings = {'ATOL': 1e-08, 'FILE_DIR': 'https://recodatasets.z20.web.core.windows.net/sarunittest/', 'TEST_USER_ID': '0003000098E85347'}
header = {'col_item': 'MovieId', 'col_rating': 'Rating', 'col_timestamp': 'Timestamp', 'col_user': 'UserId'}

    @pytest.mark.parametrize(
        "threshold,similarity_type,file",
        [
            (1, "cooccurrence", "count"),
            (1, "jaccard", "jac"),
            (1, "lift", "lift"),
            (3, "cooccurrence", "count"),
            (3, "jaccard", "jac"),
            (3, "lift", "lift"),
        ],
    )
    def test_sar_item_similarity(
        threshold, similarity_type, file, demo_usage_data, sar_settings, header
    ):
    
        model = SARSingleNode(
            similarity_type=similarity_type,
            timedecay_formula=False,
            time_decay_coefficient=30,
            threshold=threshold,
            **header
        )
    
        model.fit(demo_usage_data)
    
        true_item_similarity, row_ids, col_ids = read_matrix(
            sar_settings["FILE_DIR"] + "sim_" + file + str(threshold) + ".csv"
        )
    
        if similarity_type == "cooccurrence":
            test_item_similarity = _rearrange_to_test(
                model.item_similarity.todense(),
                row_ids,
                col_ids,
                model.item2index,
                model.item2index,
            )
            assert np.array_equal(
                true_item_similarity.astype(test_item_similarity.dtype),
                test_item_similarity,
            )
        else:
            test_item_similarity = _rearrange_to_test(
                model.item_similarity, row_ids, col_ids, model.item2index, model.item2index
            )
>           assert np.allclose(
                true_item_similarity.astype(test_item_similarity.dtype),
                test_item_similarity,
                atol=sar_settings["ATOL"],
            )
E           AssertionError: assert False
E            +  where False = <function allclose at 0x7f5b7d41d8c0>(array([[1.        , 0.        , 0.        , ..., 0.        , 0.09756098,\n        0.        ],\n       [0.        , 1.  ...  ,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.01136364, 0.        ,\n        1.        ]]), array([[1.        , 0.        , 0.        , ..., 0.        , 0.07194245,\n        0.        ],\n       [0.        , 1.  ...  ,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.00273973, 0.        ,\n        1.        ]]), atol=1e-08)
E            +    where <function allclose at 0x7f5b7d41d8c0> = np.allclose
E            +    and   array([[1.        , 0.        , 0.        , ..., 0.        , 0.09756098,\n        0.        ],\n       [0.        , 1.  ...  ,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.01136364, 0.        ,\n        1.        ]]) = <built-in method astype of numpy.ndarray object at 0x7f5b400eb090>(dtype('float64'))
E            +      where <built-in method astype of numpy.ndarray object at 0x7f5b400eb090> = array([['1', '0', '0', ..., '0', '0.0975609756097561', '0'],\n       ['0', '1', '0.0632911392405063', ..., '0', '0', '0...., '0.0609137055837563', '1',\n        '0'],\n       ['0', '0', '0', ..., '0.0113636363636364', '0', '1']], dtype='<U20').astype
E            +      and   dtype('float64') = array([[1.        , 0.        , 0.        , ..., 0.        , 0.07194245,\n        0.        ],\n       [0.        , 1.  ...  ,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.00273973, 0.        ,\n        1.        ]]).dtype

tests/unit/recommenders/models/test_sar_singlenode.py:183: AssertionError
____________________ test_sar_item_similarity[1-lift-lift] _____________________

threshold = 1, similarity_type = 'lift', file = 'lift'
demo_usage_data =                  UserId    MovieId     Timestamp  Rating
0      0003000098E85347  DQF-00358  1.433879e+09     1.0
1   ...BAE  DQF-00248  1.416292e+09     1.0
11837  00030000822E3BAE  DAF-00448  1.416292e+09     1.0

[11838 rows x 4 columns]
sar_settings = {'ATOL': 1e-08, 'FILE_DIR': 'https://recodatasets.z20.web.core.windows.net/sarunittest/', 'TEST_USER_ID': '0003000098E85347'}
header = {'col_item': 'MovieId', 'col_rating': 'Rating', 'col_timestamp': 'Timestamp', 'col_user': 'UserId'}

    @pytest.mark.parametrize(
        "threshold,similarity_type,file",
        [
            (1, "cooccurrence", "count"),
            (1, "jaccard", "jac"),
            (1, "lift", "lift"),
            (3, "cooccurrence", "count"),
            (3, "jaccard", "jac"),
            (3, "lift", "lift"),
        ],
    )
    def test_sar_item_similarity(
        threshold, similarity_type, file, demo_usage_data, sar_settings, header
    ):
    
        model = SARSingleNode(
            similarity_type=similarity_type,
            timedecay_formula=False,
            time_decay_coefficient=30,
            threshold=threshold,
            **header
        )
    
        model.fit(demo_usage_data)
    
        true_item_similarity, row_ids, col_ids = read_matrix(
            sar_settings["FILE_DIR"] + "sim_" + file + str(threshold) + ".csv"
        )
    
        if similarity_type == "cooccurrence":
            test_item_similarity = _rearrange_to_test(
                model.item_similarity.todense(),
                row_ids,
                col_ids,
                model.item2index,
                model.item2index,
            )
            assert np.array_equal(
                true_item_similarity.astype(test_item_similarity.dtype),
                test_item_similarity,
            )
        else:
            test_item_similarity = _rearrange_to_test(
                model.item_similarity, row_ids, col_ids, model.item2index, model.item2index
            )
>           assert np.allclose(
                true_item_similarity.astype(test_item_similarity.dtype),
                test_item_similarity,
                atol=sar_settings["ATOL"],
            )
E           AssertionError: assert False
E            +  where False = <function allclose at 0x7f5b7d41d8c0>(array([[0.03225806, 0.        , 0.        , ..., 0.        , 0.00346395,\n        0.        ],\n       [0.        , 0.03...41,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.00057471, 0.        ,\n        0.03448276]]), array([[1.81818182e-02, 0.00000000e+00, 0.00000000e+00, ...,\n        0.00000000e+00, 1.49644594e-03, 0.00000000e+00],\n...\n       [0.00000000e+00, 0.00000000e+00, 0.00000000e+00, ...,\n        3.69153531e-05, 0.00000000e+00, 9.70873786e-03]]), atol=1e-08)
E            +    where <function allclose at 0x7f5b7d41d8c0> = np.allclose
E            +    and   array([[0.03225806, 0.        , 0.        , ..., 0.        , 0.00346395,\n        0.        ],\n       [0.        , 0.03...41,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.00057471, 0.        ,\n        0.03448276]]) = <built-in method astype of numpy.ndarray object at 0x7f5b40e726f0>(dtype('float64'))
E            +      where <built-in method astype of numpy.ndarray object at 0x7f5b40e726f0> = array([['0.032258064516129', '0', '0', ..., '0', '0.00346395323663131',\n        '0'],\n       ['0', '0.0344827586206897...39597315', '0'],\n       ['0', '0', '0', ..., '0.000574712643678161', '0',\n        '0.0344827586206897']], dtype='<U20').astype
E            +      and   dtype('float64') = array([[1.81818182e-02, 0.00000000e+00, 0.00000000e+00, ...,\n        0.00000000e+00, 1.49644594e-03, 0.00000000e+00],\n...\n       [0.00000000e+00, 0.00000000e+00, 0.00000000e+00, ...,\n        3.69153531e-05, 0.00000000e+00, 9.70873786e-03]]).dtype

tests/unit/recommenders/models/test_sar_singlenode.py:183: AssertionError
________________ test_sar_item_similarity[3-cooccurrence-count] ________________

threshold = 3, similarity_type = 'cooccurrence', file = 'count'
demo_usage_data =                  UserId    MovieId     Timestamp  Rating
0      0003000098E85347  DQF-00358  1.433879e+09     1.0
1   ...BAE  DQF-00248  1.416292e+09     1.0
11837  00030000822E3BAE  DAF-00448  1.416292e+09     1.0

[11838 rows x 4 columns]
sar_settings = {'ATOL': 1e-08, 'FILE_DIR': 'https://recodatasets.z20.web.core.windows.net/sarunittest/', 'TEST_USER_ID': '0003000098E85347'}
header = {'col_item': 'MovieId', 'col_rating': 'Rating', 'col_timestamp': 'Timestamp', 'col_user': 'UserId'}

    @pytest.mark.parametrize(
        "threshold,similarity_type,file",
        [
            (1, "cooccurrence", "count"),
            (1, "jaccard", "jac"),
            (1, "lift", "lift"),
            (3, "cooccurrence", "count"),
            (3, "jaccard", "jac"),
            (3, "lift", "lift"),
        ],
    )
    def test_sar_item_similarity(
        threshold, similarity_type, file, demo_usage_data, sar_settings, header
    ):
    
        model = SARSingleNode(
            similarity_type=similarity_type,
            timedecay_formula=False,
            time_decay_coefficient=30,
            threshold=threshold,
            **header
        )
    
        model.fit(demo_usage_data)
    
        true_item_similarity, row_ids, col_ids = read_matrix(
            sar_settings["FILE_DIR"] + "sim_" + file + str(threshold) + ".csv"
        )
    
        if similarity_type == "cooccurrence":
            test_item_similarity = _rearrange_to_test(
                model.item_similarity.todense(),
                row_ids,
                col_ids,
                model.item2index,
                model.item2index,
            )
>           assert np.array_equal(
                true_item_similarity.astype(test_item_similarity.dtype),
                test_item_similarity,
            )
E           AssertionError: assert False
E            +  where False = <function array_equal at 0x7f5b7d41dd40>(array([[ 31.,   0.,   0., ...,   0.,  16.,   0.],\n       [  0.,  29.,   5., ...,   0.,   0.,   0.],\n       [  0.,   5.... 60.,  12.,   0.],\n       [ 16.,   0.,   0., ...,  12., 149.,   0.],\n       [  0.,   0.,   0., ...,   0.,   0.,  29.]]), matrix([[ 55.,   0.,   0., ...,   0.,  20.,   0.],\n        [  0.,  35.,   6., ...,   0.,   0.,   0.],\n        [  0.,  ...3.,  15.,   0.],\n        [ 20.,   0.,   0., ...,  15., 243.,   0.],\n        [  0.,   0.,   0., ...,   0.,   0., 103.]]))
E            +    where <function array_equal at 0x7f5b7d41dd40> = np.array_equal
E            +    and   array([[ 31.,   0.,   0., ...,   0.,  16.,   0.],\n       [  0.,  29.,   5., ...,   0.,   0.,   0.],\n       [  0.,   5.... 60.,  12.,   0.],\n       [ 16.,   0.,   0., ...,  12., 149.,   0.],\n       [  0.,   0.,   0., ...,   0.,   0.,  29.]]) = <built-in method astype of numpy.ndarray object at 0x7f5b400ea270>(dtype('float64'))
E            +      where <built-in method astype of numpy.ndarray object at 0x7f5b400ea270> = array([['31', '0', '0', ..., '0', '16', '0'],\n       ['0', '29', '5', ..., '0', '0', '0'],\n       ['0', '5', '55', ...... '12', '0'],\n       ['16', '0', '0', ..., '12', '149', '0'],\n       ['0', '0', '0', ..., '0', '0', '29']], dtype='<U4').astype
E            +      and   dtype('float64') = matrix([[ 55.,   0.,   0., ...,   0.,  20.,   0.],\n        [  0.,  35.,   6., ...,   0.,   0.,   0.],\n        [  0.,  ...3.,  15.,   0.],\n        [ 20.,   0.,   0., ...,  15., 243.,   0.],\n        [  0.,   0.,   0., ...,   0.,   0., 103.]]).dtype

tests/unit/recommenders/models/test_sar_singlenode.py:175: AssertionError
___________________ test_sar_item_similarity[3-jaccard-jac] ____________________

threshold = 3, similarity_type = 'jaccard', file = 'jac'
demo_usage_data =                  UserId    MovieId     Timestamp  Rating
0      0003000098E85347  DQF-00358  1.433879e+09     1.0
1   ...BAE  DQF-00248  1.416292e+09     1.0
11837  00030000822E3BAE  DAF-00448  1.416292e+09     1.0

[11838 rows x 4 columns]
sar_settings = {'ATOL': 1e-08, 'FILE_DIR': 'https://recodatasets.z20.web.core.windows.net/sarunittest/', 'TEST_USER_ID': '0003000098E85347'}
header = {'col_item': 'MovieId', 'col_rating': 'Rating', 'col_timestamp': 'Timestamp', 'col_user': 'UserId'}

    @pytest.mark.parametrize(
        "threshold,similarity_type,file",
        [
            (1, "cooccurrence", "count"),
            (1, "jaccard", "jac"),
            (1, "lift", "lift"),
            (3, "cooccurrence", "count"),
            (3, "jaccard", "jac"),
            (3, "lift", "lift"),
        ],
    )
    def test_sar_item_similarity(
        threshold, similarity_type, file, demo_usage_data, sar_settings, header
    ):
    
        model = SARSingleNode(
            similarity_type=similarity_type,
            timedecay_formula=False,
            time_decay_coefficient=30,
            threshold=threshold,
            **header
        )
    
        model.fit(demo_usage_data)
    
        true_item_similarity, row_ids, col_ids = read_matrix(
            sar_settings["FILE_DIR"] + "sim_" + file + str(threshold) + ".csv"
        )
    
        if similarity_type == "cooccurrence":
            test_item_similarity = _rearrange_to_test(
                model.item_similarity.todense(),
                row_ids,
                col_ids,
                model.item2index,
                model.item2index,
            )
            assert np.array_equal(
                true_item_similarity.astype(test_item_similarity.dtype),
                test_item_similarity,
            )
        else:
            test_item_similarity = _rearrange_to_test(
                model.item_similarity, row_ids, col_ids, model.item2index, model.item2index
            )
>           assert np.allclose(
                true_item_similarity.astype(test_item_similarity.dtype),
                test_item_similarity,
                atol=sar_settings["ATOL"],
            )
E           AssertionError: assert False
E            +  where False = <function allclose at 0x7f5b7d41d8c0>(array([[1.        , 0.        , 0.        , ..., 0.        , 0.09756098,\n        0.        ],\n       [0.        , 1.  ...  ,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n        1.        ]]), array([[1.        , 0.        , 0.        , ..., 0.        , 0.07194245,\n        0.        ],\n       [0.        , 1.  ...  ,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n        1.        ]]), atol=1e-08)
E            +    where <function allclose at 0x7f5b7d41d8c0> = np.allclose
E            +    and   array([[1.        , 0.        , 0.        , ..., 0.        , 0.09756098,\n        0.        ],\n       [0.        , 1.  ...  ,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n        1.        ]]) = <built-in method astype of numpy.ndarray object at 0x7f5b40b61f90>(dtype('float64'))
E            +      where <built-in method astype of numpy.ndarray object at 0x7f5b40b61f90> = array([['1', '0', '0', ..., '0', '0.0975609756097561', '0'],\n       ['0', '1', '0.0632911392405063', ..., '0', '0', '0...61', '0', '0', ..., '0.0609137055837563', '1',\n        '0'],\n       ['0', '0', '0', ..., '0', '0', '1']], dtype='<U19').astype
E            +      and   dtype('float64') = array([[1.        , 0.        , 0.        , ..., 0.        , 0.07194245,\n        0.        ],\n       [0.        , 1.  ...  ,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n        1.        ]]).dtype

tests/unit/recommenders/models/test_sar_singlenode.py:183: AssertionError
____________________ test_sar_item_similarity[3-lift-lift] _____________________

threshold = 3, similarity_type = 'lift', file = 'lift'
demo_usage_data =                  UserId    MovieId     Timestamp  Rating
0      0003000098E85347  DQF-00358  1.433879e+09     1.0
1   ...BAE  DQF-00248  1.416292e+09     1.0
11837  00030000822E3BAE  DAF-00448  1.416292e+09     1.0

[11838 rows x 4 columns]
sar_settings = {'ATOL': 1e-08, 'FILE_DIR': 'https://recodatasets.z20.web.core.windows.net/sarunittest/', 'TEST_USER_ID': '0003000098E85347'}
header = {'col_item': 'MovieId', 'col_rating': 'Rating', 'col_timestamp': 'Timestamp', 'col_user': 'UserId'}

    @pytest.mark.parametrize(
        "threshold,similarity_type,file",
        [
            (1, "cooccurrence", "count"),
            (1, "jaccard", "jac"),
            (1, "lift", "lift"),
            (3, "cooccurrence", "count"),
            (3, "jaccard", "jac"),
            (3, "lift", "lift"),
        ],
    )
    def test_sar_item_similarity(
        threshold, similarity_type, file, demo_usage_data, sar_settings, header
    ):
    
        model = SARSingleNode(
            similarity_type=similarity_type,
            timedecay_formula=False,
            time_decay_coefficient=30,
            threshold=threshold,
            **header
        )
    
        model.fit(demo_usage_data)
    
        true_item_similarity, row_ids, col_ids = read_matrix(
            sar_settings["FILE_DIR"] + "sim_" + file + str(threshold) + ".csv"
        )
    
        if similarity_type == "cooccurrence":
            test_item_similarity = _rearrange_to_test(
                model.item_similarity.todense(),
                row_ids,
                col_ids,
                model.item2index,
                model.item2index,
            )
            assert np.array_equal(
                true_item_similarity.astype(test_item_similarity.dtype),
                test_item_similarity,
            )
        else:
            test_item_similarity = _rearrange_to_test(
                model.item_similarity, row_ids, col_ids, model.item2index, model.item2index
            )
>           assert np.allclose(
                true_item_similarity.astype(test_item_similarity.dtype),
                test_item_similarity,
                atol=sar_settings["ATOL"],
            )
E           AssertionError: assert False
E            +  where False = <function allclose at 0x7f5b7d41d8c0>(array([[0.03225806, 0.        , 0.        , ..., 0.        , 0.00346395,\n        0.        ],\n       [0.        , 0.03...41,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n        0.03448276]]), array([[0.01818182, 0.        , 0.        , ..., 0.        , 0.00149645,\n        0.        ],\n       [0.        , 0.02...23,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n        0.00970874]]), atol=1e-08)
E            +    where <function allclose at 0x7f5b7d41d8c0> = np.allclose
E            +    and   array([[0.03225806, 0.        , 0.        , ..., 0.        , 0.00346395,\n        0.        ],\n       [0.        , 0.03...41,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n        0.03448276]]) = <built-in method astype of numpy.ndarray object at 0x7f5b3f8e9f30>(dtype('float64'))
E            +      where <built-in method astype of numpy.ndarray object at 0x7f5b3f8e9f30> = array([['0.032258064516129', '0', '0', ..., '0', '0.00346395323663131',\n        '0'],\n       ['0', '0.0344827586206897...9463',\n        '0.00671140939597315', '0'],\n       ['0', '0', '0', ..., '0', '0', '0.0344827586206897']], dtype='<U20').astype
E            +      and   dtype('float64') = array([[0.01818182, 0.        , 0.        , ..., 0.        , 0.00149645,\n        0.        ],\n       [0.        , 0.02...23,\n        0.        ],\n       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n        0.00970874]]).dtype

tests/unit/recommenders/models/test_sar_singlenode.py:183: AssertionError

miguelgfierro · 2022-01-21T12:55:52Z

I think @simonzhaoms was looking into removing drop_duplicates, should I close this PR?

simonzhaoms · 2022-02-23T10:08:32Z

@miguelgfierro I am working on this PR and will close this PR after I fix the failures in tests.

miguelgfierro

LGTM

miguelgfierro · 2022-02-28T09:01:56Z

@simonzhaoms is this PR ready to be merged?

simonzhaoms · 2022-02-28T09:50:00Z

Yes. @miguelgfierro

Remove drop_duplicates() from SAR method fix #1464

6aa7d8d

miguelgfierro requested review from anargyri, gramhagen, loomlike and wutaomsft as code owners December 17, 2021 09:24

flake is complaining

562d19b

anargyri approved these changes Jan 21, 2022

View reviewed changes

Merge branch 'staging' into miguelgfierro-patch-2

f1d53a7

simonzhaoms added 4 commits February 23, 2022 16:36

Merge branch 'staging' into miguelgfierro-patch-2

bc76330

Typos

9f9d497

Define self.unity_user_affinity inside __init__()

8392535

Remove drop_duplicates() from SAR method

0c4d707

simonzhaoms added 3 commits February 24, 2022 14:57

Remove duplicates in testing data

19b8fa3

Remove duplicates in test data for recommend_k_items

557edcc

Allow duplicates in score data

eb5294a

miguelgfierro commented Feb 28, 2022

View reviewed changes

simonzhaoms merged commit 96b5053 into staging Feb 28, 2022

miguelgfierro deleted the miguelgfierro-patch-2 branch February 28, 2022 14:34

danb27 mentioned this pull request Mar 9, 2022

danb27/sar_single_node_improvements: allow for getting most frequent … #1666

Merged

4 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Remove drop_duplicates() from SAR method fix #1464 #1588

Remove drop_duplicates() from SAR method fix #1464 #1588

miguelgfierro commented Dec 17, 2021

miguelgfierro commented Dec 17, 2021

miguelgfierro commented Jan 21, 2022

simonzhaoms commented Feb 23, 2022 •

edited

Loading

miguelgfierro left a comment

miguelgfierro commented Feb 28, 2022

simonzhaoms commented Feb 28, 2022

Remove drop_duplicates() from SAR method fix #1464 #1588

Remove drop_duplicates() from SAR method fix #1464 #1588

Conversation

miguelgfierro commented Dec 17, 2021

Description

Related Issues

Checklist:

miguelgfierro commented Dec 17, 2021

miguelgfierro commented Jan 21, 2022

simonzhaoms commented Feb 23, 2022 • edited Loading

miguelgfierro left a comment

Choose a reason for hiding this comment

miguelgfierro commented Feb 28, 2022

simonzhaoms commented Feb 28, 2022

simonzhaoms commented Feb 23, 2022 •

edited

Loading