Change dimension 'nI' for model 'maxout' from 768 to 192

[rishkum]

Hi

I am getting the following error:

  File "C:\Anaconda3\envs\prodigy117\lib\site-packages\thinc\layers\maxout.py", line 74, in init
    model.set_dim("nI", get_width(X))
  File "C:\Anaconda3\envs\prodigy117\lib\site-packages\thinc\model.py", line 192, in set_dim
    raise ValueError(err
ValueError: Attempt to change dimension 'nI' for model 'maxout' from 768 to 192.

Config file for textcat:

[paths]
train = null
dev = null
vectors = "en_core_web_md"
init_tok2vec = "output/model20.bin"
raw_text = "data.jsonl"

[system]
seed = 0
gpu_allocator = null

[nlp]
lang = "en"
pipeline = ["tok2vec","textcat_multilabel"]
disabled = []
before_creation = null
after_creation = null
after_pipeline_creation = null
batch_size = 1000
tokenizer = {"@tokenizers":"spacy.Tokenizer.v1"}

[components]

[components.textcat_multilabel]
factory = "textcat_multilabel"
scorer = {"@scorers":"spacy.textcat_multilabel_scorer.v1"}
threshold = 0.5

[components.textcat_multilabel.model]
@architectures = "spacy.TextCatEnsemble.v2"
nO = null

[components.textcat_multilabel.model.linear_model]
@architectures = "spacy.TextCatBOW.v2"
exclusive_classes = false
ngram_size = 1
no_output_layer = false
nO = null

[components.textcat_multilabel.model.tok2vec]
@architectures = "spacy.Tok2Vec.v2"

[components.textcat_multilabel.model.tok2vec.embed]
@architectures = "spacy.MultiHashEmbed.v1"
width = 64
rows = [2000,2000,1000,1000,1000,1000]
attrs = ["ORTH","LOWER","PREFIX","SUFFIX","SHAPE","ID"]
include_static_vectors = true

[components.textcat_multilabel.model.tok2vec.encode]
@architectures = "spacy.MaxoutWindowEncoder.v2"
width = 256
depth = 8
window_size = 1
maxout_pieces = 3

[components.tok2vec]
factory = "tok2vec"

[components.tok2vec.model]
@architectures = "spacy.Tok2Vec.v2"

[components.tok2vec.model.embed]
@architectures = "spacy.MultiHashEmbed.v2"
width = ${components.tok2vec.model.encode.width}
attrs = ["ORTH","SHAPE"]
rows = [5000,2500]
include_static_vectors = true

[components.tok2vec.model.encode]
@architectures = "spacy.MaxoutWindowEncoder.v2"
width = 256
depth = 8
window_size = 1
maxout_pieces = 3

[corpora]

[corpora.dev]
@readers = "spacy.Corpus.v1"
path = ${paths.dev}
gold_preproc = false
max_length = 0
limit = 0
augmenter = null

[corpora.train]
@readers = "spacy.Corpus.v1"
path = ${paths.train}
gold_preproc = false
max_length = 0
limit = 0
augmenter = null

[training]
seed = ${system.seed}
gpu_allocator = ${system.gpu_allocator}
dropout = 0.1
accumulate_gradient = 1
patience = 1000
max_epochs = 0
max_steps = 2000
eval_frequency = 100
frozen_components = []
dev_corpus = "corpora.dev"
train_corpus = "corpora.train"
before_to_disk = null
annotating_components = []

[training.batcher]
@batchers = "spacy.batch_by_words.v1"
discard_oversize = false
tolerance = 0.2
get_length = null

[training.batcher.size]
@schedules = "compounding.v1"
start = 100
stop = 1000
compound = 1.001
t = 0.0

[training.logger]
@loggers = "spacy.ConsoleLogger.v1"
progress_bar = false

[training.optimizer]
@optimizers = "Adam.v1"
beta1 = 0.9
beta2 = 0.999
L2_is_weight_decay = true
L2 = 0.01
grad_clip = 1.0
use_averages = false
eps = 0.00000001
learn_rate = 0.001

[training.score_weights]
cats_score = 1.0
cats_score_desc = null
cats_micro_p = null
cats_micro_r = null
cats_micro_f = null
cats_macro_p = null
cats_macro_r = null
cats_macro_f = null
cats_macro_auc = null
cats_f_per_type = null
cats_macro_auc_per_type = null


[initialize]
vectors = ${paths.vectors}
init_tok2vec = ${paths.init_tok2vec}
vocab_data = null
lookups = null
before_init = null
after_init = null

[initialize.components]

[initialize.components.textcat_multilabel]

[initialize.components.textcat_multilabel.labels]
@readers = "spacy.read_labels.v1"
path = "assets\\labels\\textcat_multilabel.json"
require = false

[initialize.tokenizer]

Spacy: 3.2.2
Python: Python 3.7.12

[pmbaumgartner]

Hey @rishkum, thanks for starting a discussion.

In your config, you have the line

init_tok2vec = "output/model20.bin"

Where did this model come from? I'm guessing the dimensions of those vectors are different than the dimensions that this architecture is expecting as input. Typically you'll point this value to vectors that have been through pretraining, or in some cases an existing model (which seems like what you have).