LIMA
Libre Multilingual Analyzer — C++ API
Loading...
Searching...
No Matches
train_lemmatization.cpp File Reference
#include <string>
#include <unordered_map>
#include <algorithm>
#include "conllu/treebank.h"
#include "train_lemmatization.h"
#include "morph_model/morph_model_builder.h"
#include "deeplima/utils/str_index.h"
#include "static_graph/dict.h"
#include "deeplima/torch_wrp/torch_matrix.h"
#include "tasks/lemmatization/model/seq2seq_for_lemmatization.h"
#include <boost/functional/hash.hpp>
#include "deeplima/utils/backport.h"
#include "deeplima/utils/split_string.h"

Go to the source code of this file.

Classes

struct  deeplima::lemmatization::train::form_morph_t
 
struct  deeplima::lemmatization::train::form_morph_hash
 

Namespaces

namespace  deeplima
 
namespace  deeplima::lemmatization
 
namespace  deeplima::lemmatization::train
 

Typedefs

typedef unordered_map< form_morph_t, unordered_map< StringIndex::idx_t, size_t >, form_morph_hash > deeplima::lemmatization::train::form2lemma_t
 

Functions

bool deeplima::lemmatization::train::operator== (const form_morph_t &a, const form_morph_t &b)
 
DictsHolder deeplima::lemmatization::train::build_char_dicts (const form2lemma_t &form2lemma, StringIndex &str_idx)
 
set< morph_model::feat_base_t > deeplima::lemmatization::train::find_fixed (const morph_model::morph_model_t &lang_morph_model, const form2lemma_t &form2lemma, StringIndex &str_idx)
 
void deeplima::lemmatization::train::vectorize_dataset (const morph_model::morph_model_t &lang_morph_model, const form2lemma_t &form2lemma, StringIndex &str_idx, const DictsHolder dh, uint32_t max_len, vector< TorchMatrix< int64_t > > &v_seq_input, vector< vector< TorchMatrix< int64_t > > > &v_cat_input, vector< TorchMatrix< int64_t > > &v_gold)
 
int deeplima::lemmatization::train::train_lemmatization (const train_params_lemmatization_t &params)
 

Variables

const char32_t deeplima::lemmatization::train::START = 0x10FFFE
 
const char32_t deeplima::lemmatization::train::EOS = 0x10FFFF