File size: 1,849 Bytes
cd123bf |
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 |
from dataclasses import dataclass
@dataclass
class DatasetConfig:
"""
Configuration class for dataset generation parameters
Attributes:
embedding_dim: Dimension for embedding layer output
train_size: Number of samples in training set
val_size: Number of samples in validation set
test_size: Number of samples in test set
random_state: Random seed for reproducibility
min_word_freq: Minimum word frequency to include in vocabulary
load_from_disk: Load dataset from local dir. If false download from huggin face
path_to_data: Path to local dataset data
max_seq_len: Maximum sequence length (will be padded/truncated to this)
lowercase: Whether to convert text to lowercase
remove_punct: Whether to remove punctuation
pad_token: Padding token
unk_token: Unknown token
"""
embedding_dim: int = 64
train_size: int = 10000
val_size: int = 5000
test_size: int = 5000
random_state: int = 42
min_word_freq: int = 1
load_from_disk: bool = False
path_to_data: str = "./datasets"
max_seq_len: int = 300
lowercase: bool = True
remove_punct: bool = False
pad_token: str = "<PAD>"
unk_token: str = "<UNK>"
@dataclass
class TextProcessorConfig:
"""
Configuration class for text processor parameters (params should be equal dataset config)
Attributes:
max_seq_len: Maximum sequence length (will be padded/truncated to this)
lowercase: Whether to convert text to lowercase
remove_punct: Whether to remove punctuation
pad_token: Padding token
unk_token: Unknown token
"""
max_seq_len: int = 300
lowercase: bool = True
remove_punct: bool = False
pad_token: str = "<PAD>"
unk_token: str = "<UNK>"
|