# coding=utf-8
# Copyright 2023 the Falcon authors and HuggingFace Inc. team. 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.
""" Falcon configuration"""
from transformers.configuration_utils import PretrainedConfig
from transformers.utils import logging
from transformers import AutoConfig
logger = logging.get_logger(__name__)
class MAELMConfig(PretrainedConfig):
"""
This is the configuration class to store the configuration of a [`FalconModel`]. It is used to instantiate a Falcon
model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
defaults will yield a similar configuration to that of the
[tiiuae/falcon-7b](https://huggingface.co/tiiuae/falcon-7b) architecture.
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
documentation from [`PretrainedConfig`] for more information.
Args:
vocab_size (`int`, *optional*, defaults to 65024):
Vocabulary size of the Falcon model. Defines the number of different tokens that can be represented by the
`inputs_ids` passed when calling [`FalconModel`]
hidden_size (`int`, *optional*, defaults to 4544):
Dimension of the hidden representations.
num_hidden_layers (`int`, *optional*, defaults to 32):
Number of hidden layers in the Transformer decoder.
num_attention_heads (`int`, *optional*, defaults to 71):
Number of attention heads for each attention layer in the Transformer encoder.
layer_norm_epsilon (`float`, *optional*, defaults to 1e-05):
The epsilon used by the layer normalization layers.
initializer_range (`float`, *optional*, defaults to 0.02):
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
use_cache (`bool`, *optional*, defaults to `True`):
Whether the model should return the last key/values attentions (not used by all models). Only relevant if
`config.is_decoder=True`.
hidden_dropout (`float`, *optional*, defaults to 0.0):
The dropout probability for MLP layers.
attention_dropout (`float`, *optional*, defaults to 0.0):
The dropout probability for attention layers.
num_kv_heads (`int`, *optional*):
Number of key-value heads to use per attention layer. If unset, defaults to the same value as
`num_attention_heads`.
alibi (`bool`, *optional*, defaults to `False`):
Whether to use ALiBi positional biases during self-attention.
new_decoder_architecture (`bool`, *optional*, defaults to `False`):
Whether to use the new (Falcon-40B) decoder architecture. If `True`, the `multi_query` and `parallel_attn`
arguments are ignored, as the new decoder always uses parallel attention.
multi_query (`bool`, *optional*, defaults to `True`):
Whether to use multi-query attention in the decoder. Ignored when `new_decoder_architecture` is `True`.
parallel_attn (`bool`, *optional*, defaults to `True`):
Whether to compute attention in parallel with the feedforward layer. If False, they are consecutive
instead, as in the original Transformer architecture. Ignored when `new_decoder_architecture` is `True`.
bias (`bool`, *optional*, defaults to `False`):
Whether to use bias on Linear layers.
max_position_embeddings (`int`, *optional*, defaults to 2048):
The maximum sequence length that this model might ever be used with, when `alibi` is `False`. Pretrained
Falcon models with RoPE support up to 2048 tokens.
rope_theta (`float`, *optional*, defaults to 10000.0):
The base period of the RoPE embeddings.
rope_scaling (`Dict`, *optional*):
Dictionary containing the scaling configuration for the RoPE embeddings. Currently supports two scaling
strategies: linear and dynamic. Their scaling factor must be a float greater than 1. The expected format is
`{"type": strategy name, "factor": scaling factor}`. When using this flag, don't update
`max_position_embeddings` to the expected new maximum. See the following thread for more information on how
these scaling strategies behave:
https://www.reddit.com/r/LocalLLaMA/comments/14mrgpr/dynamically_scaled_rope_further_increases/. This is an
experimental feature, subject to breaking API changes in future versions.
bos_token_id (`int`, *optional*, defaults to 11):
The id of the "beginning-of-sequence" token.
eos_token_id (`int`, *optional*, defaults to 11):
The id of the "end-of-sequence" token.
"""
model_type = "MAELM"
def __init__(
self,
seed=42,
cache_dir=None,
do_train=True,
do_eval=False,
do_test=False,
dataset_name=None,
spect_len=2992,
train_dataset_list=[{'train_file': '/mnt/bn/music-nas-dxj1/datasets/MCC_AIGC/mccaigc_train_1w.csv', \
'train_tokenized_data': None, 'train_data_root': '/mnt/bn/music-nas-dxj1/datasets/MCC_AIGC/logmel',}],
per_device_eval_batch_size=32,
preprocessing_num_workers=64,
overwrite_cache=True,
output_dir='/mnt/bn/music-nas-dxj1/VWork/ckpts_vault/cap_lynx-apm_umg_PT-mccaigc1w_FT',
save_interval_steps=1000,
overwrite_output_dir=True,
gradient_accumulation_steps=1,
num_train_epochs=50,
per_device_train_batch_size=12,
learning_rate=0.00005,
lm_lr_ratio=0.1,
tokenizer_name='meta-llama/Llama-2-7b-hf',
resume_from_checkpoint=None,
resume_from_pth='epoch_4-step_8639-allstep_60000.pth',
backbone={'name': 'MAEViT', 'arch': 'b', 'patch_size': 16, 'mask_ratio': 0.0, 'img_size': [80, 2992], \
'ckpt': 'epoch_20.pth'},
neck={'name': 'LMDecoder', 'patch_size': 16, 'img_size': [80, 2992], 'in_chans': 3, 'embed_dim': 768, \
'decoder_embed_dim': 4544, 'freeze_decoder': True, 'decoder_type': 'meta-llama/Llama-2-7b-hf'},
wandb={'proj': 'ATRena_cap', 'expname': 'cap_lynx_apmPT_mccaigc1wFT'},
**kwargs,
):
self.backbone = backbone
self.neck = neck
self.tokenizer_name = tokenizer_name
self._name_or_path = None
self.resume_from_checkpoint = resume_from_checkpoint
self.resume_from_pth = resume_from_pth
self.auto_map = {"AutoConfig": "configuration_maelm.MAELMConfig",
"AutoModel": "modeling_maelm.MAEForCausalLM"}