How to use "spacy-transformers.TransformerModel.v3" in my present NER transformer training?

[karndeepsingh]

Hi,
I have been training NER transformer model and I have been using version 2 of transformer i.e "spacy-transformers.TransformerModel.v2" and when I am trying to replace this version 2 transformer to version 3 transformer i.e "spacy-transformers.TransformerModel.v3" in my present config transformer file, it throws an error. How I can use this version 3 on my current training for NER.
I am using spacy-3.1.2. Please help me to incorporate this latest version of transformer into my current training for NER. Following is my current config file with version 2 transformer.

[paths]
train = null
dev = null
raw = null
init_tok2vec = null

[system]
seed = 342
gpu_allocator = "pytorch"

#"relation_extractor"
[nlp]
lang = "en"
pipeline = ["transformer","ner","relation_extractor"]
disabled = []
before_creation = null
after_creation = null
after_pipeline_creation = null
tokenizer = {"@tokenizers":"spacy.Tokenizer.v1"}
batch_size = 2

[components]

[components.transformer]
factory = "transformer"
max_batch_items = 4096
set_extra_annotations = {"@annotation_setters":"spacy-transformers.null_annotation_setter.v1"}

[components.transformer.model]
@architectures = "spacy-transformers.TransformerModel.v1"
name = "roberta-base"
tokenizer_config = {"use_fast": true}
# get_spans = 32

[components.transformer.model.get_spans]
@span_getters = "spacy-transformers.strided_spans.v1"
window = 128
stride = 96

[components.ner]
factory = "ner"

[components.ner.model]
@architectures = "spacy.TransitionBasedParser.v2"
state_type = "ner"
extra_state_tokens = false
hidden_width = 64
maxout_pieces = 2
use_upper = false
nO = null

[components.ner.model.tok2vec]
@architectures = "spacy-transformers.TransformerListener.v1"
grad_factor = 1.0

[components.ner.model.tok2vec.pooling]
@layers = "reduce_mean.v1"

[components.relation_extractor]
factory = "relation_extractor"
threshold = 0.5

[components.relation_extractor.model]
@architectures = "rel_model.v1"

[components.relation_extractor.model.create_instance_tensor]
@architectures = "rel_instance_tensor.v1"

[components.relation_extractor.model.create_instance_tensor.tok2vec]
@architectures = "spacy-transformers.TransformerListener.v1"
grad_factor = 1.0

[components.relation_extractor.model.create_instance_tensor.tok2vec.pooling]
@layers = "reduce_mean.v1"

[components.relation_extractor.model.create_instance_tensor.pooling]
@layers = "reduce_mean.v1"

[components.relation_extractor.model.create_instance_tensor.get_instances]
@misc = "rel_instance_generator.v1"
max_length = 30

[components.relation_extractor.model.classification_layer]
@architectures = "rel_classification_layer.v1"
nI = null
nO = null

[initialize]

[initialize.components]

[corpora]

[corpora.dev]
@readers = "Gold_ents_Corpus.v1"
file = ${paths.dev}
# # # max_length = 512
# [corpora.dev]
# @readers = "spacy.Corpus.v1"
# path = ${paths.dev}
# gold_preproc = True
# max_length = 512


[corpora.train]
@readers = "Gold_ents_Corpus.v1"
file = ${paths.train}
# # max_length = 512

# [corpora.train]
# @readers = "spacy.Corpus.v1"
# path = ${paths.train}
# max_length = 100

# [corpora.dev]
# @readers = "spacy.Corpus.v1"
# path = ${paths.dev}
# max_length = 0

[training]
seed = ${system.seed}
gpu_allocator = ${system.gpu_allocator}
dropout = 0.1
accumulate_gradient = 1
patience = 15000
max_epochs = 0
max_steps = 15000
eval_frequency = 500
frozen_components = []
dev_corpus = "corpora.dev"
train_corpus = "corpora.train"
before_to_disk = null
logger = {"@loggers":"spacy.ConsoleLogger.v1"}

# [training.batcher]
# @batchers = "spacy.batch_by_padded.v1"
# discard_oversize = true
# size = 2000
# buffer = 256

[training.batcher]
@batchers = "spacy.batch_by_padded.v1"
discard_oversize = true
size = 512
buffer = 512
# [training.batcher]
# @batchers = "spacy.batch_by_sequence.v1"
# size = 1
# get_length = null

[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

[training.optimizer.learn_rate]
@schedules = "warmup_linear.v1"
warmup_steps = 250
total_steps = 20000
initial_rate = 5e-5

[training.score_weights]
rel_micro_p = 0.0
rel_micro_r = 0.0
rel_micro_f = 1.0

[adrianeboyd]

spacy-transformers.TransformerModel.v3 was first published in spacy-transformers v1.1.0, so you probably just need to upgrade spacy-transformers.

(There was never an official release with spacy-transformers.TransformerModel.v2 as the default version. Did you install it from source at some point? You might want to try uninstalling spacy-transformers several times to be sure no traces are left and reinstalling from scratch, since sometimes pip doesn't clean up old installs properly.)

[karndeepsingh]

@adrianeboyd Thanks for the reply. I installed it from the conda commands provided on spacy installation page.
Can you please guide me the difference between the V3 transformer model and V2 Transformer model?

[adrianeboyd]

Please take a look at the docs: https://spacy.io/api/architectures#transformers