configuration_ctrl.py 5.63 KB
Newer Older
keskarnitish's avatar
keskarnitish committed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
# coding=utf-8
# Copyright 2018 Salesforce and HuggingFace Inc. team.
# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
""" Salesforce CTRL configuration """


import logging

from .configuration_utils import PretrainedConfig

Aymeric Augustin's avatar
Aymeric Augustin committed
22

keskarnitish's avatar
keskarnitish committed
23
24
25
26
logger = logging.getLogger(__name__)

CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP = {"ctrl": "https://storage.googleapis.com/sf-ctrl/pytorch/ctrl-config.json"}

27

keskarnitish's avatar
keskarnitish committed
28
class CTRLConfig(PretrainedConfig):
Lysandre Debut's avatar
Lysandre Debut committed
29
30
31
32
    """
        This is the configuration class to store the configuration of an :class:`~transformers.CTRLModel`.
        It is used to instantiate an CTRL model according to the specified arguments, defining the model
        architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of
Lysandre's avatar
Lysandre committed
33
        the `ctrl <https://huggingface.co/ctrl>`__ architecture from SalesForce.
Lysandre Debut's avatar
Lysandre Debut committed
34
35
36
37

        Configuration objects inherit from  :class:`~transformers.PretrainedConfig` and can be used
        to control the model outputs. Read the documentation from  :class:`~transformers.PretrainedConfig`
        for more information.
keskarnitish's avatar
keskarnitish committed
38

Lysandre Debut's avatar
Lysandre Debut committed
39
40
41
42
43
44
45
46
        Args:
            vocab_size (:obj:`int`, optional, defaults to 246534):
                Vocabulary size of the CTRL model. Defines the different tokens that
                can be represented by the `inputs_ids` passed to the forward method of :class:`~transformers.CTRLModel`.
            n_positions (:obj:`int`, optional, defaults to 256):
                The maximum sequence length that this model might ever be used with.
                Typically set this to something large just in case (e.g., 512 or 1024 or 2048).
            n_ctx (:obj:`int`, optional, defaults to 256):
47
                Dimensionality of the causal mask (usually same as n_positions).
Lysandre Debut's avatar
Lysandre Debut committed
48
49
50
            n_embd (:obj:`int`, optional, defaults to 1280):
                Dimensionality of the embeddings and hidden states.
            dff (:obj:`int`, optional, defaults to 8192):
51
                Dimensionality of the inner dimension of the FFN.
Lysandre Debut's avatar
Lysandre Debut committed
52
53
54
55
56
57
58
59
60
61
62
63
64
65
            n_layer (:obj:`int`, optional, defaults to 48):
                Number of hidden layers in the Transformer encoder.
            n_head (:obj:`int`, optional, defaults to 16):
                Number of attention heads for each attention layer in the Transformer encoder.
            resid_pdrop (:obj:`float`, optional, defaults to 0.1):
                The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
            embd_pdrop (:obj:`int`, optional, defaults to 0.1):
                The dropout ratio for the embeddings.
            attn_pdrop (:obj:`float`, optional, defaults to 0.1):
                The dropout ratio for the attention.
            layer_norm_epsilon (:obj:`float`, optional, defaults to 1e-6):
                The epsilon to use in the layer normalization layers
            initializer_range (:obj:`float`, optional, defaults to 0.02):
                The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
Lysandre's avatar
Lysandre committed
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82

        Example::

            from transformers import CTRLModel, CTRLConfig

            # Initializing a CTRL configuration
            configuration = CTRLConfig()

            # Initializing a model from the configuration
            model = CTRLModel(configuration)

            # Accessing the model configuration
            configuration = model.config

        Attributes:
            pretrained_config_archive_map (Dict[str, str]):
                A dictionary containing all the available pre-trained checkpoints.
keskarnitish's avatar
keskarnitish committed
83
    """
84

keskarnitish's avatar
keskarnitish committed
85
    pretrained_config_archive_map = CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP
Julien Chaumond's avatar
Julien Chaumond committed
86
    model_type = "ctrl"
keskarnitish's avatar
keskarnitish committed
87
88
89

    def __init__(
        self,
thomwolf's avatar
thomwolf committed
90
        vocab_size=246534,
thomwolf's avatar
thomwolf committed
91
92
        n_positions=256,
        n_ctx=256,
keskarnitish's avatar
keskarnitish committed
93
94
95
96
97
98
99
100
101
        n_embd=1280,
        dff=8192,
        n_layer=48,
        n_head=16,
        resid_pdrop=0.1,
        embd_pdrop=0.1,
        attn_pdrop=0.1,
        layer_norm_epsilon=1e-6,
        initializer_range=0.02,
102
        summary_type="cls_index",
keskarnitish's avatar
keskarnitish committed
103
104
105
106
107
108
        summary_use_proj=True,
        summary_activation=None,
        summary_proj_to_labels=True,
        summary_first_dropout=0.1,
        **kwargs
    ):
Julien Chaumond's avatar
Julien Chaumond committed
109
        super().__init__(**kwargs)
thomwolf's avatar
thomwolf committed
110
        self.vocab_size = vocab_size
thomwolf's avatar
thomwolf committed
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
        self.n_ctx = n_ctx
        self.n_positions = n_positions
        self.n_embd = n_embd
        self.n_layer = n_layer
        self.n_head = n_head
        self.dff = dff
        self.resid_pdrop = resid_pdrop
        self.embd_pdrop = embd_pdrop
        self.attn_pdrop = attn_pdrop
        self.layer_norm_epsilon = layer_norm_epsilon
        self.initializer_range = initializer_range

        self.summary_type = summary_type
        self.summary_use_proj = summary_use_proj
        self.summary_activation = summary_activation
        self.summary_first_dropout = summary_first_dropout
        self.summary_proj_to_labels = summary_proj_to_labels
keskarnitish's avatar
keskarnitish committed
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143

    @property
    def max_position_embeddings(self):
        return self.n_positions

    @property
    def hidden_size(self):
        return self.n_embd

    @property
    def num_attention_heads(self):
        return self.n_head

    @property
    def num_hidden_layers(self):
        return self.n_layer