Вещественные числа и задача о задачах

RobertDeMilo

Oct 21st, 2023

Never

Add comment

Not a member of Pastebin yet? Sign Up, it unlocks many cool features!

C++ 8.48 KB | None | 0 0

raw download clone embed print report

#include <algorithm>
#include <cmath>
#include <iostream>
#include <map>
#include <set>
#include <string>
#include <utility>
#include <vector>
using namespace std;
const int MAX_RESULT_DOCUMENT_COUNT = 5;
const double EPSILON = 1e-6;
struct Document {
int id;
double relevance;
int rating;
};
enum class DocumentStatus
{
ACTUAL,
IRRELEVANT,
BANNED,
REMOVED
};
string ReadLine() {
string s;
getline(cin, s);
return s;
}
int ReadLineWithNumber() {
int result;
cin >> result;
ReadLine();
return result;
}
vector<string> SplitIntoWords(const string& text) {
vector<string> words;
string word;
for (const char c : text) {
if (c == ' ') {
words.push_back(word);
word = "";
}
else {
word += c;
}
}
words.push_back(word);
return words;
}
class SearchServer {
public:
void SetStopWords(const string& text) {
for (const string& word : SplitIntoWords(text)) {
stop_words_.insert(word);
}
}
void AddDocument(int document_id, const string& document, DocumentStatus status, const vector<int>& ratings) {
const vector<string> words = SplitIntoWordsNoStop(document);
const double inv_word_count = 1.0 / words.size();
for (const string& word : words) {
word_to_document_freqs_[word][document_id] += inv_word_count;
}
document_ratings_[document_id] = { ComputeAverageRating(ratings), status };
}
int GetDocumentCount()
{
return static_cast<int>(document_ratings_.size());
}
tuple<vector<string>, DocumentStatus> MatchDocument(const string& raw_query, int document_id) const
{
// Если документ не соответствует запросу(нет пересечений по плюс - словам или есть минус - слово), вектор слов нужно вернуть пустым.
Query query = ParseQuery(raw_query);
vector<string> pluss_wordss;
for (const string& word : query.plus_words)
{
if (word_to_document_freqs_.count(word) != 0 && word_to_document_freqs_.at(word).count(document_id) != 0)
{
pluss_wordss.push_back(word);
}
}
for (const string& word : query.minus_words)
{
if (word_to_document_freqs_.count(word) != 0 && word_to_document_freqs_.at(word).count(document_id) != 0)
{
pluss_wordss.clear();
break;
}
}
// map<int, Doc> document_ratings_;
return { pluss_wordss, document_ratings_.at(document_id).status };
}
vector<Document> FindTopDocuments(const string& raw_query, DocumentStatus statuss = DocumentStatus::ACTUAL) const {
const Query query = ParseQuery(raw_query);
auto matched_documents2 = FindAllDocuments(query);
vector<Document> matched_documents;
for (auto it = find_if(matched_documents2.begin(), matched_documents2.end(),
[this, statuss](Document& document)
{return (document_ratings_.count(document.id) != 0 && document_ratings_.at(document.id).status == statuss); }); it != matched_documents2.end();
it = find_if(next(it), matched_documents2.end(),
[this, statuss](Document& document)
{return (document_ratings_.count(document.id) != 0 && document_ratings_.at(document.id).status == statuss); }))
{
matched_documents.push_back(*it);
}
sort(matched_documents.begin(), matched_documents.end(),
[](const Document& lhs, const Document& rhs) {
if (abs(lhs.relevance - rhs.relevance) > EPSILON)
{
return lhs.rating > rhs.rating;
}
return lhs.relevance > rhs.relevance;
});
if (matched_documents.size() > MAX_RESULT_DOCUMENT_COUNT) {
matched_documents.resize(MAX_RESULT_DOCUMENT_COUNT);
}
return matched_documents;
}
private:
set<string> stop_words_;
map<string, map<int, double>> word_to_document_freqs_;
struct Query {
set<string> plus_words;
set<string> minus_words;
};
struct QueryWord {
string data;
bool is_minus;
bool is_stop;
};
struct Doc
{
int rating;
DocumentStatus status;
};
// id , {rat, status}
map<int, Doc> document_ratings_;
bool IsStopWord(const string& word) const {
return stop_words_.count(word) > 0;
}
vector<string> SplitIntoWordsNoStop(const string& text) const {
vector<string> words;
for (const string& word : SplitIntoWords(text)) {
if (!IsStopWord(word)) {
words.push_back(word);
}
}
return words;
}
static int ComputeAverageRating(const vector<int>& ratings) {
int rating_sum = 0;
for (const int rating : ratings) {
rating_sum += rating;
}
return rating_sum / static_cast<int>(ratings.size());
}
QueryWord ParseQueryWord(string text) const {
bool is_minus = false;
// Word shouldn't be empty
if (text[0] == '-') {
is_minus = true;
text = text.substr(1);
}
return {
text,
is_minus,
IsStopWord(text)
};
}
Query ParseQuery(const string& text) const {
Query query;
for (const string& word : SplitIntoWords(text)) {
const QueryWord query_word = ParseQueryWord(word);
if (!query_word.is_stop) {
if (query_word.is_minus) {
query.minus_words.insert(query_word.data);
}
else {
query.plus_words.insert(query_word.data);
}
}
}
return query;
}
// Existence required
double ComputeWordInverseDocumentFreq(const string& word) const {
return log(document_ratings_.size() * 1.0 / word_to_document_freqs_.at(word).size());
}
vector<Document> FindAllDocuments(const Query& query) const {
map<int, double> document_to_relevance;
for (const string& word : query.plus_words) {
if (word_to_document_freqs_.count(word) == 0) {
continue;
}
const double inverse_document_freq = ComputeWordInverseDocumentFreq(word);
for (const auto [document_id, term_freq] : word_to_document_freqs_.at(word)) {
document_to_relevance[document_id] += term_freq * inverse_document_freq;
}
}
for (const string& word : query.minus_words) {
if (word_to_document_freqs_.count(word) == 0) {
continue;
}
for (const auto [document_id, _] : word_to_document_freqs_.at(word)) {
document_to_relevance.erase(document_id);
}
}
vector<Document> matched_documents;
for (const auto [document_id, relevance] : document_to_relevance) {
matched_documents.push_back({
document_id,
relevance,
document_ratings_.at(document_id).rating
});
}
return matched_documents;
}
};
void PrintDocument(const Document& document) {
cout << "{ "s
<< "document_id = "s << document.id << ", "s
<< "relevance = "s << document.relevance << ", "s
<< "rating = "s << document.rating
<< " }"s << endl;
}
int main() {
SearchServer search_server;
search_server.SetStopWords("и в на"s);
search_server.AddDocument(0, "белый кот и модный ошейник"s, DocumentStatus::ACTUAL, { 8, -3 });
search_server.AddDocument(1, "пушистый кот пушистый хвост"s, DocumentStatus::ACTUAL, { 7, 2, 7 });
search_server.AddDocument(2, "ухоженный пёс выразительные глаза"s, DocumentStatus::ACTUAL, { 5, -12, 2, 1 });
for (const Document& document : search_server.FindTopDocuments("ухоженный кот"s)) {
PrintDocument(document);
}
}

Add Comment

Please, Sign In to add comment