Hugging Face's logo

Spaces:
nvidia
/
Plan2Align-NV
Sleeping

App Files Community
1
Plan2Align-NV
File size: 2,892 Bytes 
2aebc50
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
 
// Copyright 2016 Google Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//     http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.!
#include "pretokenizer_for_training.h"

#include "testharness.h"
#include "third_party/absl/strings/str_cat.h"
#include "third_party/absl/strings/str_join.h"
#include "third_party/absl/strings/str_split.h"
#include "trainer_interface.h"

namespace sentencepiece {
namespace pretokenizer {

class MockPretokenizer : public PretokenizerForTrainingInterface {
 public:
  MockPretokenizer() {}
  ~MockPretokenizer() {}

  SentencePieceText Tokenize(absl::string_view text) const override {
    return spt_;
  }

  util::Status status() const override { return util::OkStatus(); }

  void SetOutput(const SentencePieceText &spt) { spt_ = spt; }

 private:
  SentencePieceText spt_;
};

TEST(PretokenizerForTrainingTest, BaseTest) {
  MockPretokenizer mock;

  {
    SentencePieceText spt;
    spt.set_text("I love sentencepiece");
    auto *p1 = spt.add_pieces();
    p1->set_surface("I");
    p1->set_begin(0);
    p1->set_end(1);

    auto *p2 = spt.add_pieces();
    p2->set_surface("love");
    p2->set_begin(2);
    p2->set_end(6);

    auto *p3 = spt.add_pieces();
    p3->set_surface("sentence");
    p3->set_begin(7);
    p3->set_end(15);

    auto *p4 = spt.add_pieces();
    p4->set_surface("piece");
    p4->set_begin(15);
    p4->set_end(20);

    mock.SetOutput(spt);

    const auto expected =
        absl::StrCat("I", TrainerInterface::kWSStr, "love",
                     TrainerInterface::kWSStr, "sentence||||piece");
    EXPECT_EQ(expected,
              absl::StrJoin(mock.PreTokenize("I love sentencepiece"), "||||"));
  }

  {
    SentencePieceText spt;
    spt.set_text("γ“γ‚Œγ―γƒšγƒ³γ§γ™");
    auto *p1 = spt.add_pieces();
    p1->set_surface("γ“γ‚Œ");
    p1->set_begin(0);
    p1->set_end(6);

    auto *p2 = spt.add_pieces();
    p2->set_surface("は");
    p2->set_begin(6);
    p2->set_end(9);

    auto *p3 = spt.add_pieces();
    p3->set_surface("γƒšγƒ³");
    p3->set_begin(9);
    p3->set_end(15);

    auto *p4 = spt.add_pieces();
    p4->set_surface("です");
    p4->set_begin(15);
    p4->set_end(21);

    mock.SetOutput(spt);

    const auto expected = "γ“γ‚Œ||||は||||γƒšγƒ³||||です";
    EXPECT_EQ(expected,
              absl::StrJoin(mock.PreTokenize("γ“γ‚Œγ―γƒšγƒ³γ§γ™"), "||||"));
  }
}

}  // namespace pretokenizer
}  // namespace sentencepiece