Mercurial > hgrepos > Python2 > PyMuPDF

comparison mupdf-source/thirdparty/tesseract/src/training/unicharset/lstmtester.h @ 2:b50eed0cc0ef upstream

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
ADD: MuPDF v1.26.7: the MuPDF source as downloaded by a default build of PyMuPDF 1.26.4. The directory name has changed: no version number in the expanded directory now.
author Franz Glasner <fzglas.hg@dom66.de>
date Mon, 15 Sep 2025 11:43:07 +0200
parents
children
comparison
equal deleted inserted replaced
1:1d09e1dec1d9 2:b50eed0cc0ef
1 ///////////////////////////////////////////////////////////////////////
2 // File: lstmtester.h
3 // Description: Top-level line evaluation class for LSTM-based networks.
4 // Author: Ray Smith
5 //
6 // (C) Copyright 2016, Google Inc.
7 // Licensed under the Apache License, Version 2.0 (the "License");
8 // you may not use this file except in compliance with the License.
9 // You may obtain a copy of the License at
10 // http://www.apache.org/licenses/LICENSE-2.0
11 // Unless required by applicable law or agreed to in writing, software
12 // distributed under the License is distributed on an "AS IS" BASIS,
13 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14 // See the License for the specific language governing permissions and
15 // limitations under the License.
16 ///////////////////////////////////////////////////////////////////////
17
18 #ifndef TESSERACT_TRAINING_LSTMTESTER_H_
19 #define TESSERACT_TRAINING_LSTMTESTER_H_
20
21 #include "export.h"
22
23 #include "lstmtrainer.h"
24
25 #include <mutex>
26 #include <string>
27 #include <vector>
28
29 namespace tesseract {
30
31 class TESS_UNICHARSET_TRAINING_API LSTMTester {
32 public:
33 LSTMTester(int64_t max_memory);
34
35 // Loads a set of lstmf files that were created using the lstm.train config to
36 // tesseract into memory ready for testing. Returns false if nothing was
37 // loaded. The arg is a filename of a file that lists the filenames, with one
38 // name per line. Conveniently, tesstrain.py generates such a file, along
39 // with the files themselves.
40 bool LoadAllEvalData(const char *filenames_file);
41 // Loads a set of lstmf files that were created using the lstm.train config to
42 // tesseract into memory ready for testing. Returns false if nothing was
43 // loaded.
44 bool LoadAllEvalData(const std::vector<std::string> &filenames);
45
46 // Runs an evaluation asynchronously on the stored eval data and returns a
47 // string describing the results of the previous test. Args match TestCallback
48 // declared in lstmtrainer.h:
49 // iteration: Current learning iteration number.
50 // training_errors: If not null, is an array of size ET_COUNT, indexed by
51 // the ErrorTypes enum and indicates the current errors measured by the
52 // trainer, and this is a serious request to run an evaluation. If null,
53 // then the caller is just polling for the results of the previous eval.
54 // model_data: is the model to evaluate, which should be a serialized
55 // LSTMTrainer.
56 // training_stage: an arbitrary number on the progress of training.
57 std::string RunEvalAsync(int iteration, const double *training_errors,
58 const TessdataManager &model_mgr, int training_stage);
59 // Runs an evaluation synchronously on the stored eval data and returns a
60 // string describing the results. Args as RunEvalAsync, except verbosity,
61 // which outputs errors, if 1, or all results if 2.
62 std::string RunEvalSync(int iteration, const double *training_errors, const TessdataManager &model_mgr,
63 int training_stage, int verbosity);
64
65 private:
66 // Helper thread function for RunEvalAsync.
67 // LockIfNotRunning must have returned true before calling ThreadFunc, and
68 // it will call UnlockRunning to release the lock after RunEvalSync completes.
69 void ThreadFunc();
70 // Returns true if there is currently nothing running, and takes the lock
71 // if there is nothing running.
72 bool LockIfNotRunning();
73 // Releases the running lock.
74 void UnlockRunning();
75
76 // The data to test with.
77 DocumentCache test_data_;
78 int total_pages_ = 0;
79 // Flag that indicates an asynchronous test is currently running.
80 // Protected by running_mutex_.
81 bool async_running_ = false;
82 std::mutex running_mutex_;
83 // Stored copies of the args for use while running asynchronously.
84 int test_iteration_ = 0;
85 const double *test_training_errors_ = nullptr;
86 TessdataManager test_model_mgr_;
87 int test_training_stage_ = 0;
88 std::string test_result_;
89 };
90
91 } // namespace tesseract
92
93 #endif // TESSERACT_TRAINING_LSTMTESTER_H_