Hi there,
I am currently stuck trying to load a custom ner model trained using spacy 3.2.3 and spacy-transformers 1.1.6.
When loading the model into prodigy (prodigy-1.11.6-cp39-cp39-linux_x86_64.whl), I get the following error:
ner_correct_compose-prodigy | catalogue.RegistryError: [E893] Could not find function 'spacy-transformers.Tok2VecTransformer.v3' in function registry 'architectures'. If you're using a custom function, make sure the code is available. If the function is provided by a third-party package, e.g. spacy-transformers, make sure the package is installed in your environment.
ner_correct_compose-prodigy |
ner_correct_compose-prodigy | Available names: spacy-legacy.CharacterEmbed.v1, spacy-legacy.EntityLinker.v1, spacy-legacy.HashEmbedCNN.v1, spacy-legacy.MaxoutWindowEncoder.v1, spacy-legacy.MishWindowEncoder.v1, spacy-legacy.MultiHashEmbed.v1, spacy-legacy.Tagger.v1, spacy-legacy.TextCatBOW.v1, spacy-legacy.TextCatCNN.v1, spacy-legacy.TextCatEnsemble.v1, spacy-legacy.Tok2Vec.v1, spacy-legacy.TransitionBasedParser.v1, spacy-transformers.Tok2VecTransformer.v1, spacy-transformers.TransformerListener.v1, spacy-transformers.TransformerModel.v1, spacy.CharacterEmbed.v2, spacy.EntityLinker.v1, spacy.HashEmbedCNN.v2, spacy.MaxoutWindowEncoder.v2, spacy.MishWindowEncoder.v2, spacy.MultiHashEmbed.v2, spacy.PretrainCharacters.v1, spacy.PretrainVectors.v1, spacy.SpanCategorizer.v1, spacy.Tagger.v1, spacy.TextCatBOW.v2, spacy.TextCatCNN.v2, spacy.TextCatEnsemble.v2, spacy.TextCatLowData.v1, spacy.Tok2Vec.v2, spacy.Tok2VecListener.v1, spacy.TorchBiLSTMEncoder.v1, spacy.TransitionBasedParser.v2
despite the fact that I my training config shows:
[components.transformer.model]
@architectures = "spacy-transformers.TransformerModel.v1"
name = "roberta-base"
I am running Prodigy within a docker container which has python 3.9, spacy 3.1.3 (hopefully not causing issues for the model trained with 3.2.3), and spacy-transformers 1.1.4 installed.
Please help 
See full error and config below:
ner_correct_compose-prodigy | warnings.warn(warn_msg)
ner_correct_compose-prodigy | Traceback (most recent call last):
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/runpy.py", line 197, in _run_module_as_main
ner_correct_compose-prodigy | return _run_code(code, main_globals, None,
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/runpy.py", line 87, in _run_code
ner_correct_compose-prodigy | exec(code, run_globals)
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/prodigy/__main__.py", line 61, in <module>
ner_correct_compose-prodigy | controller = recipe(*args, use_plac=True)
ner_correct_compose-prodigy | File "cython_src/prodigy/core.pyx", line 329, in prodigy.core.recipe.recipe_decorator.recipe_proxy
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/plac_core.py", line 367, in call
ner_correct_compose-prodigy | cmd, result = parser.consume(arglist)
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/plac_core.py", line 232, in consume
ner_correct_compose-prodigy | return cmd, self.func(*(args + varargs + extraopts), **kwargs)
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/prodigy/recipes/ner.py", line 215, in correct
ner_correct_compose-prodigy | nlp = spacy.load(spacy_model)
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/spacy/__init__.py", line 51, in load
ner_correct_compose-prodigy | return util.load_model(
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/spacy/util.py", line 349, in load_model
ner_correct_compose-prodigy | return load_model_from_path(Path(name), **kwargs)
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/spacy/util.py", line 415, in load_model_from_path
ner_correct_compose-prodigy | nlp = load_model_from_config(config, vocab=vocab, disable=disable, exclude=exclude)
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/spacy/util.py", line 452, in load_model_from_config
ner_correct_compose-prodigy | nlp = lang_cls.from_config(
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/spacy/language.py", line 1714, in from_config
ner_correct_compose-prodigy | nlp.add_pipe(
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/spacy/language.py", line 776, in add_pipe
ner_correct_compose-prodigy | pipe_component = self.create_pipe(
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/spacy/language.py", line 660, in create_pipe
ner_correct_compose-prodigy | resolved = registry.resolve(cfg, validate=validate)
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/thinc/config.py", line 729, in resolve
ner_correct_compose-prodigy | resolved, _ = cls._make(
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/thinc/config.py", line 778, in _make
ner_correct_compose-prodigy | filled, _, resolved = cls._fill(
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/thinc/config.py", line 833, in _fill
ner_correct_compose-prodigy | filled[key], validation[v_key], final[key] = cls._fill(
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/thinc/config.py", line 833, in _fill
ner_correct_compose-prodigy | filled[key], validation[v_key], final[key] = cls._fill(
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/thinc/config.py", line 832, in _fill
ner_correct_compose-prodigy | promise_schema = cls.make_promise_schema(value, resolve=resolve)
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/thinc/config.py", line 1023, in make_promise_schema
ner_correct_compose-prodigy | func = cls.get(reg_name, func_name)
ner_correct_compose-prodigy | File "/usr/local/lib/python3.9/site-packages/spacy/util.py", line 136, in get
ner_correct_compose-prodigy | raise RegistryError(
ner_correct_compose-prodigy | catalogue.RegistryError: [E893] Could not find function 'spacy-transformers.Tok2VecTransformer.v3' in function registry 'architectures'. If you're using a custom function, make sure the code is available. If the function is provided by a third-party package, e.g. spacy-transformers, make sure the package is installed in your environment.
ner_correct_compose-prodigy |
ner_correct_compose-prodigy | Available names: spacy-legacy.CharacterEmbed.v1, spacy-legacy.EntityLinker.v1, spacy-legacy.HashEmbedCNN.v1, spacy-legacy.MaxoutWindowEncoder.v1, spacy-legacy.MishWindowEncoder.v1, spacy-legacy.MultiHashEmbed.v1, spacy-legacy.Tagger.v1, spacy-legacy.TextCatBOW.v1, spacy-legacy.TextCatCNN.v1, spacy-legacy.TextCatEnsemble.v1, spacy-legacy.Tok2Vec.v1, spacy-legacy.TransitionBasedParser.v1, spacy-transformers.Tok2VecTransformer.v1, spacy-transformers.TransformerListener.v1, spacy-transformers.TransformerModel.v1, spacy.CharacterEmbed.v2, spacy.EntityLinker.v1, spacy.HashEmbedCNN.v2, spacy.MaxoutWindowEncoder.v2, spacy.MishWindowEncoder.v2, spacy.MultiHashEmbed.v2, spacy.PretrainCharacters.v1, spacy.PretrainVectors.v1, spacy.SpanCategorizer.v1, spacy.Tagger.v1, spacy.TextCatBOW.v2, spacy.TextCatCNN.v2, spacy.TextCatEnsemble.v2, spacy.TextCatLowData.v1, spacy.Tok2Vec.v2, spacy.Tok2VecListener.v1, spacy.TorchBiLSTMEncoder.v1, spacy.TransitionBasedParser.v2
[paths]
train = null
dev = null
vectors = null
init_tok2vec = null
[system]
gpu_allocator = "pytorch"
seed = 0
[nlp]
lang = "en"
pipeline = ["transformer","tagger","parser","ner"]
batch_size = 128
disabled = []
before_creation = null
after_creation = null
after_pipeline_creation = null
tokenizer = {"@tokenizers":"spacy.Tokenizer.v1"}
[components]
[components.ner]
factory = "ner"
incorrect_spans_key = null
moves = null
scorer = {"@scorers":"spacy.ner_scorer.v1"}
update_with_oracle_cut_size = 100
[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
pooling = {"@layers":"reduce_mean.v1"}
upstream = "*"
[components.parser]
source = "en_core_web_trf"
replace_listeners = ["model.tok2vec"]
[components.tagger]
source = "en_core_web_trf"
replace_listeners = ["model.tok2vec"]
[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"
[components.transformer.model.get_spans]
@span_getters = "spacy-transformers.strided_spans.v1"
window = 128
stride = 96
[components.transformer.model.tokenizer_config]
use_fast = true
[corpora]
[corpora.dev]
@readers = "spacy.Corpus.v1"
path = ${paths.dev}
max_length = 0
gold_preproc = false
limit = 0
augmenter = null
[corpora.train]
@readers = "spacy.Corpus.v1"
path = ${paths.train}
max_length = 500
gold_preproc = false
limit = 0
augmenter = null
[training]
accumulate_gradient = 3
dev_corpus = "corpora.dev"
train_corpus = "corpora.train"
seed = ${system.seed}
gpu_allocator = ${system.gpu_allocator}
dropout = 0.1
patience = 1600
max_epochs = 0
max_steps = 20000
eval_frequency = 200
frozen_components = []
before_to_disk = null
annotating_components = []
[training.batcher]
@batchers = "spacy.batch_by_padded.v1"
discard_oversize = true
size = 2000
buffer = 256
get_length = null
[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
[training.optimizer.learn_rate]
@schedules = "warmup_linear.v1"
warmup_steps = 250
total_steps = 20000
initial_rate = 0.00005
[training.score_weights]
tag_acc = 0.33
dep_uas = 0.17
dep_las = 0.17
dep_las_per_type = null
sents_p = null
sents_r = null
sents_f = 0.0
ents_f = 0.33
ents_p = 0.0
ents_r = 0.0
ents_per_type = null
[pretraining]
[initialize]
vectors = null
init_tok2vec = ${paths.init_tok2vec}
vocab_data = null
lookups = null
before_init = null
after_init = null
[initialize.components]
[initialize.tokenizer]