### Install RusStress with Pip

Source: https://github.com/mashapo/russtress/blob/master/README.rst

Install the RusStress library using pip. This is the simplest way to get started.

```bash
pip install russtress
```

--------------------------------

### Process Files with RusStress

Source: https://context7.com/mashapo/russtress/llms.txt

Read text from files, apply stress marks, and save the output. Includes an example for line-by-line processing of large files.

```python
from russtress import Accent

accent = Accent()

# Read input file
with open('input.txt', 'r', encoding='utf-8') as f:
    text = f.read()

# Add stress marks
stressed_text = accent.put_stress(text)

# Write output file
with open('output_stressed.txt', 'w', encoding='utf-8') as f:
    f.write(stressed_text)

# Process line by line for large files
with open('large_input.txt', 'r', encoding='utf-8') as infile:
    with open('large_output.txt', 'w', encoding='utf-8') as outfile:
        for line in infile:
            stressed_line = accent.put_stress(line)
            outfile.write(stressed_line)
```

--------------------------------

### Preserve Punctuation and Formatting with put_stress

Source: https://context7.com/mashapo/russtress/llms.txt

The `put_stress` method preserves original formatting, punctuation, and capitalization when adding stress marks. This example shows stress being added to a sentence with a period.

```python
from russtress import Accent

accent = Accent()

# Sentence with punctuation preserved
sentence = 'Мама мыла раму.'
print(accent.put_stress(sentence))
# Output: "Ма'ма мы'ла ра'му."
```

--------------------------------

### Custom Category Tokenization

Source: https://context7.com/mashapo/russtress/llms.txt

The `tokenize` function allows defining custom character categories for segmentation. This example specifies categories for digits, whitespace, and Cyrillic characters.

```python
from russtress.tokenizer import tokenize

# Custom category tokenization
custom_categories = [
    '0123456789',
    ' 	
',
    'абвгдеёжзийклмнопрстуфхцчшщъыьэюяАБВГДЕЁЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯ'
]
tokens = tokenize('Тест 123', categories=custom_categories)
print(tokens)
# Output: ['Тест', ' ', '123']
```

--------------------------------

### Tokenization Preserving All Characters

Source: https://context7.com/mashapo/russtress/llms.txt

The `tokenize` function from `russtress.tokenizer` preserves all characters, including numbers and punctuation, when segmenting text. This example demonstrates tokenizing a sentence with numbers.

```python
from russtress.tokenizer import tokenize

# Tokenization preserves all characters
text_with_numbers = 'В 2024 году будет 366 дней.'
tokens = tokenize(text_with_numbers)
print(tokens)
# Output: ['В', ' ', '2024', ' ', 'году', ' ', 'будет', ' ', '366', ' ', 'дней', '.']
```

--------------------------------

### Use Custom Stress Symbol

Source: https://context7.com/mashapo/russtress/llms.txt

The `put_stress` method allows specifying a custom stress symbol. This example uses the combining acute accent character (U+0301) instead of an apostrophe.

```python
from russtress import Accent

accent = Accent()

# Custom stress symbol
text = 'Привет мир'
accented = accent.put_stress(text, stress_symbol='́')  # Using combining acute accent
print(accented)
# Output: "Приве́т ми́р"
```

--------------------------------

### Initialize Accent Model and Stress a Phrase

Source: https://context7.com/mashapo/russtress/llms.txt

Initialize the Accent model by importing the Accent class and creating an instance. Then, use the `put_stress` method to add stress marks to a Russian phrase. The output will have stress marks inserted after the stressed vowel.

```python
from russtress import Accent

# Initialize the accent model (loads LSTM weights)
accent = Accent()

# Basic usage - stress a simple phrase
text = 'Проставь, пожалуйста, ударения'
accented_text = accent.put_stress(text)
print(accented_text)
# Output: "Проста'вь, пожа'луйста, ударе'ния"
```

--------------------------------

### Basic Tokenization

Source: https://context7.com/mashapo/russtress/llms.txt

Import the `tokenize` function from `russtress.tokenizer` to split Russian text into tokens. This function preserves character categories like digits, spaces, punctuation, and Cyrillic characters.

```python
from russtress.tokenizer import tokenize

# Basic tokenization
text = 'Привет, мир! Как дела?'
tokens = tokenize(text)
print(tokens)
# Output: ['Привет', ',', ' ', 'мир', '!', ' ', 'Как', ' ', 'дела', '?']
```

--------------------------------

### Tokenizer Module

Source: https://context7.com/mashapo/russtress/llms.txt

Provides text segmentation functionality that splits Russian text into tokens while preserving character categories.

```APIDOC
## [FUNCTION] tokenize

### Description
Splits a string into a list of tokens based on character categories.

### Parameters
#### Request Body
- **text** (string) - Required - The text to tokenize.
- **categories** (list) - Optional - Custom character categories for tokenization.

### Response
#### Success Response (200)
- **tokens** (list) - A list of strings representing the segmented text.

### Request Example
{
  "text": "Привет, мир!"
}

### Response Example
{
  "tokens": ["Привет", ",", " ", "мир", "!"]
}
```

--------------------------------

### Batch Process Multiple Texts with RusStress

Source: https://context7.com/mashapo/russtress/llms.txt

Reuse a single Accent instance to process multiple strings efficiently without reloading the model.

```python
from russtress import Accent

# Initialize once, use many times
accent = Accent()

texts = [
    'Доброе утро!',
    'Как поживаете?',
    'Спасибо, хорошо.',
    'До свидания!'
]

# Process batch of texts
stressed_texts = [accent.put_stress(text) for text in texts]
for original, stressed in zip(texts, stressed_texts):
    print(f'{original} -> {stressed}')

# Output:
# Доброе утро! -> До'брое у'тро!
# Как поживаете? -> Как пожива'ете?
# Спасибо, хорошо. -> Спаси'бо, хорошо'.
# До свидания! -> До свида'ния!
```

--------------------------------

### Accent.put_stress Method

Source: https://context7.com/mashapo/russtress/llms.txt

The primary method for adding stress marks to Russian text. It preserves original formatting, punctuation, and capitalization while adding stress annotations to words with two or more syllables.

```APIDOC
## [METHOD] put_stress

### Description
Adds stress marks to Russian text using an LSTM model. It considers word context to handle homographs and preserves original text structure.

### Parameters
#### Request Body
- **text** (string) - Required - The Russian text to process.
- **stress_symbol** (string) - Optional - The symbol used to mark stress (default is an apostrophe).

### Response
#### Success Response (200)
- **result** (string) - The input text with stress marks inserted after stressed vowels.

### Request Example
{
  "text": "Мама мыла раму.",
  "stress_symbol": "'"
}

### Response Example
{
  "result": "Ма'ма мы'ла ра'му."
}
```

--------------------------------

### Stress Multi-Sentence Paragraphs

Source: https://context7.com/mashapo/russtress/llms.txt

The `put_stress` method can handle multi-sentence paragraphs, preserving newlines and punctuation while adding stress marks to words with two or more syllables.

```python
from russtress import Accent

accent = Accent()

# Multi-sentence paragraph
paragraph = '''Русский язык — один из восточнославянских языков.
Он является одним из наиболее распространённых языков мира.'''
print(accent.put_stress(paragraph))
# Output preserves newlines and punctuation with stress marks added
```

--------------------------------

### Apply Stress Marks to Text

Source: https://github.com/mashapo/russtress/blob/master/README.rst

Use the Accent class from the russtress library to automatically add stress marks to Russian text. Import the Accent class and instantiate it, then call the put_stress method with your text.

```python
from russtress import Accent
accent = Accent()
text = 'Проставь, пожалуйста, ударения'
accented_text = accent.put_stress(text)
accented_text
```

--------------------------------

### Check if Word is Small (Monosyllabic)

Source: https://context7.com/mashapo/russtress/llms.txt

The `is_small` utility function determines if a word has one or fewer syllables. Words identified as 'small' will not have stress marks applied by the `put_stress` method.

```python
from russtress.utils import is_small

# Check if word has 1 or fewer syllables (won't be stressed)
print(is_small('я'))      # True - single vowel
print(is_small('дом'))    # True - single vowel
print(is_small('мама'))   # False - two vowels (two syllables)
print(is_small('привет')) # False - two vowels (two syllables)
```

--------------------------------

### Parse Text into Words

Source: https://context7.com/mashapo/russtress/llms.txt

Use the `parse_the_phrase` utility function to parse text into words, marking clause boundaries. The output is a list of lowercase words and empty strings, with '_' indicating clause breaks.

```python
from russtress.utils import parse_the_phrase

# Parse text into words with clause boundaries marked
text = 'Привет, мир! Как дела?'
words = parse_the_phrase(text)
print(words)
# Output: ['привет', '', '_', '', 'мир', '', '_', '', 'как', '', 'дела', '', '_', '']
```

--------------------------------

### Stress a Single Word

Source: https://context7.com/mashapo/russtress/llms.txt

Use the `put_stress` method of the Accent class to add stress marks to a single Russian word. The method returns the word with an apostrophe after the stressed vowel.

```python
from russtress import Accent

accent = Accent()

# Single word stress
word = 'молоко'
print(accent.put_stress(word))
# Output: "молоко'"
```

--------------------------------

### Check if Word is in Dictionary

Source: https://context7.com/mashapo/russtress/llms.txt

The `is_in_dictionary` utility function checks if a given word exists in the stress prediction system's dictionary. It returns `True` if the word is found, `False` otherwise.

```python
from russtress.utils import is_in_dictionary

# Check if a word is in the stress dictionary
print(is_in_dictionary('молоко'))
# Output: True or False depending on dictionary contents
```

--------------------------------

### Context-Dependent Stress Prediction

Source: https://context7.com/mashapo/russtress/llms.txt

The LSTM model uses word context to predict stress, handling homographs. Words with two or more syllables receive stress marks, while monosyllabic words do not.

```python
from russtress import Accent

accent = Accent()

# The model considers context for stress prediction
# Words with 2+ syllables get stress marks
text1 = 'Красивые мечты сбываются'
print(accent.put_stress(text1))
# Output: "Краси'вые мечты' сбыва'ются"
```

```python
from russtress import Accent

accent = Accent()

# Short words (1 syllable) are not stressed
text2 = 'Я иду в лес'
print(accent.put_stress(text2))
# Output: "Я иду' в лес"  (only 'иду' gets stress - others are monosyllabic)
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.