github.com/weaviate/weaviate@v1.24.6/modules/text2vec-jinaai/vectorizer/texts_test.go (about)

     1  //                           _       _
     2  // __      _____  __ ___   ___  __ _| |_ ___
     3  // \ \ /\ / / _ \/ _` \ \ / / |/ _` | __/ _ \
     4  //  \ V  V /  __/ (_| |\ V /| | (_| | ||  __/
     5  //   \_/\_/ \___|\__,_| \_/ |_|\__,_|\__\___|
     6  //
     7  //  Copyright © 2016 - 2024 Weaviate B.V. All rights reserved.
     8  //
     9  //  CONTACT: hello@weaviate.io
    10  //
    11  
    12  package vectorizer
    13  
    14  import (
    15  	"context"
    16  	"testing"
    17  
    18  	"github.com/stretchr/testify/assert"
    19  	"github.com/stretchr/testify/require"
    20  )
    21  
    22  // as used in the nearText searcher
    23  func TestVectorizingTexts(t *testing.T) {
    24  	type testCase struct {
    25  		name                string
    26  		input               []string
    27  		expectedJinaAIModel string
    28  		jinaAIModel         string
    29  	}
    30  
    31  	tests := []testCase{
    32  		{
    33  			name:                "single word",
    34  			input:               []string{"hello"},
    35  			jinaAIModel:         "jina-embedding-v2",
    36  			expectedJinaAIModel: "jina-embedding-v2",
    37  		},
    38  		{
    39  			name:                "multiple words",
    40  			input:               []string{"hello world, this is me!"},
    41  			jinaAIModel:         "jina-embedding-v2",
    42  			expectedJinaAIModel: "jina-embedding-v2",
    43  		},
    44  		{
    45  			name:                "multiple sentences (joined with a dot)",
    46  			input:               []string{"this is sentence 1", "and here's number 2"},
    47  			jinaAIModel:         "jina-embedding-v2",
    48  			expectedJinaAIModel: "jina-embedding-v2",
    49  		},
    50  		{
    51  			name:                "multiple sentences already containing a dot",
    52  			input:               []string{"this is sentence 1.", "and here's number 2"},
    53  			jinaAIModel:         "jina-embedding-v2",
    54  			expectedJinaAIModel: "jina-embedding-v2",
    55  		},
    56  		{
    57  			name:                "multiple sentences already containing a question mark",
    58  			input:               []string{"this is sentence 1?", "and here's number 2"},
    59  			jinaAIModel:         "jina-embedding-v2",
    60  			expectedJinaAIModel: "jina-embedding-v2",
    61  		},
    62  		{
    63  			name:                "multiple sentences already containing an exclamation mark",
    64  			input:               []string{"this is sentence 1!", "and here's number 2"},
    65  			jinaAIModel:         "jina-embedding-v2",
    66  			expectedJinaAIModel: "jina-embedding-v2",
    67  		},
    68  		{
    69  			name:                "multiple sentences already containing comma",
    70  			input:               []string{"this is sentence 1,", "and here's number 2"},
    71  			jinaAIModel:         "jina-embedding-v2",
    72  			expectedJinaAIModel: "jina-embedding-v2",
    73  		},
    74  	}
    75  
    76  	for _, test := range tests {
    77  		t.Run(test.name, func(t *testing.T) {
    78  			client := &fakeClient{}
    79  
    80  			v := New(client)
    81  
    82  			settings := &fakeClassConfig{
    83  				jinaAIModel: test.jinaAIModel,
    84  			}
    85  			vec, err := v.Texts(context.Background(), test.input, settings)
    86  
    87  			require.Nil(t, err)
    88  			assert.Equal(t, []float32{0.1, 1.1, 2.1, 3.1}, vec)
    89  			assert.Equal(t, test.input, client.lastInput)
    90  			assert.Equal(t, client.lastConfig.Model, test.expectedJinaAIModel)
    91  		})
    92  	}
    93  }