SpanCategorizer prediction issue

[smisra1]

I successfully trained the spancat model. However I can't get a predicted span with a label form the same. I have added the code I used for prediction. How could I get the result as the chosen span and the label associated with it

nlp = spacy.load('output_/model-best')
rest = 'red riding hood'
spancat = nlp.get_pipe("spancat")
doc1 = nlp(rest)
scores = spancat.predict(docs=[doc1])
print (scores)
print(spancat.set_annotations(docs=[doc1], indices_scores=scores))
print(doc1.spans)

output:

(Ragged(data=array([[0, 1],
       [1, 2],
       [2, 3],
       [0, 2],
       [1, 3],
       [0, 3]], dtype=int32), lengths=array([6], dtype=int32), data_shape=(-1, 2), cumsums=None), array([[1.7509257e-04, 3.9194015e-04, 8.2362228e-04, 6.4050657e-04,
        3.7456930e-04, 2.5173824e-03, 1.7261749e-06, 2.5276720e-06,
        1.9942349e-06, 7.7653231e-06, 3.7047492e-05, 2.7342865e-05,
        6.1166734e-06, 5.1169744e-05, 1.1258147e-05, 7.5871399e-06,
        1.6391847e-05, 1.1133462e-05, 4.7089416e-06, 8.2503793e-06,
        5.3636755e-05, 1.2530764e-06, 5.6583604e-06, 5.3173004e-05,
        5.7688494e-06, 4.5809170e-06, 7.0639817e-06, 5.6532199e-06,
        3.2455771e-06, 1.9376210e-05, 9.1620159e-06, 8.7704093e-06,
        1.1047254e-05, 4.9847768e-06, 1.1122912e-05, 3.0112710e-06],
       [3.2811097e-04, 1.2592147e-06, 6.8819511e-04, 5.1525119e-04,
        1.4796455e-05, 4.4062387e-04, 2.4169108e-07, 3.4838678e-07,
        7.4420637e-07, 2.0216121e-06, 2.4065614e-06, 3.1049724e-06,
        3.8556684e-07, 3.3903874e-05, 1.5230102e-06, 2.5984725e-06,
        9.0027152e-07, 3.9645843e-06, 8.5195211e-07, 9.8862949e-07,
        9.4052001e-07, 2.2608310e-06, 8.0877726e-07, 2.6583614e-06,
        1.2487329e-06, 1.6210024e-06, 2.9016226e-06, 5.5132750e-07,
        5.9044538e-07, 1.0386509e-06, 4.7527263e-07, 5.7237645e-07,
        6.8771294e-07, 1.4301764e-06, 9.8684154e-07, 1.1001401e-06],
       [1.9582402e-04, 1.2306891e-05, 7.6035671e-05, 1.6316431e-04,
        3.6198089e-05, 9.6696615e-04, 2.1253497e-06, 1.3608022e-06,
        2.0224431e-06, 2.1414819e-06, 3.7015582e-06, 2.9562968e-06,
        5.6866031e-07, 4.1470710e-05, 1.5456051e-06, 3.8880694e-06,
        2.9160171e-06, 5.6454992e-06, 1.0828620e-06, 1.1927226e-06,
        1.4717371e-06, 4.6429009e-06, 3.1273671e-06, 6.4621474e-07,
        1.5405590e-06, 2.4284959e-06, 4.9434389e-06, 9.9949250e-07,
        1.0101946e-06, 1.3998633e-06, 1.2815768e-06, 1.1263324e-06,
        1.5267084e-06, 1.2419976e-06, 1.6613166e-06, 1.6968447e-06],
       [1.6711043e-04, 2.6113929e-03, 8.1074795e-05, 5.2020478e-05,
        8.1885897e-05, 6.8006193e-04, 9.9166812e-07, 1.3885983e-07,
        2.9603875e-06, 2.6340063e-06, 5.7517536e-06, 6.3746206e-06,
        3.7670350e-07, 3.2674870e-05, 2.2663507e-06, 1.9334341e-06,
        2.8930738e-06, 2.5053216e-06, 1.2757396e-06, 7.2127864e-07,
        3.4121701e-06, 1.4163888e-06, 2.0214657e-06, 2.4234641e-06,
        1.2618582e-06, 2.5072982e-06, 3.2205749e-06, 8.6074351e-07,
        1.4756525e-06, 2.3843834e-06, 1.4132477e-06, 1.2444625e-06,
        1.8460257e-06, 7.9173975e-07, 1.1545985e-06, 1.0458251e-06],
       [2.2760409e-04, 8.2496399e-06, 1.3827188e-04, 1.3278225e-04,
        1.0984882e-05, 3.6088473e-04, 1.0850412e-06, 6.4568621e-07,
        1.0007650e-06, 1.4730794e-06, 2.9248688e-06, 2.6272701e-06,
        3.5280857e-07, 4.8570135e-05, 1.1370971e-06, 2.5190354e-06,
        1.6944030e-06, 3.4841344e-06, 7.5244049e-07, 8.1960815e-07,
        8.4280185e-07, 3.3897845e-06, 1.4170899e-06, 7.4523973e-07,
        1.1359666e-06, 1.3057773e-06, 3.1176655e-06, 5.5839854e-07,
        5.2113677e-07, 1.0034322e-06, 7.5004897e-07, 5.3164337e-07,
        6.6009238e-07, 9.3623424e-07, 7.1181370e-07, 1.1186901e-06],
       [1.8647860e-04, 8.5528368e-01, 3.1492009e-04, 1.3644088e-04,
        2.6609478e-04, 6.4349332e-04, 3.6643774e-04, 1.2785018e-06,
        2.9186216e-05, 3.1340853e-05, 6.7809946e-05, 4.9155664e-05,
        8.3009754e-06, 2.4522917e-04, 2.8938224e-05, 2.3798468e-05,
        3.7578669e-05, 3.6067871e-05, 1.1233010e-05, 1.3270917e-05,
        2.1508893e-05, 1.4216518e-05, 2.4917350e-05, 3.6635279e-06,
        1.9339970e-05, 2.6067619e-05, 3.4147593e-05, 1.5430720e-05,
        3.2560241e-05, 1.3830537e-05, 3.3103141e-05, 1.5422480e-05,
        2.0607637e-05, 1.4141647e-05, 8.2574479e-06, 2.5114598e-05]],
      dtype=float32))
None
{'sc': [red riding hood]}

My config.cfg is below, if required:

[paths]
train = "./spansdata.spacy"
dev = "./spansdata.spacy"
vectors = null
init_tok2vec = null

[system]
gpu_allocator = null
seed = 0

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

[components]

[components.spancat]
factory = "spancat"
max_positive = null
scorer = {"@scorers":"spacy.spancat_scorer.v1"}
spans_key = "sc"
threshold = 0.5

[components.spancat.model]
@architectures = "spacy.SpanCategorizer.v1"

[components.spancat.model.reducer]
@layers = "spacy.mean_max_reducer.v1"
hidden_size = 128

[components.spancat.model.scorer]
@layers = "spacy.LinearLogistic.v1"
nO = null
nI = null

[components.spancat.model.tok2vec]
@architectures = "spacy.Tok2Vec.v1"

[components.spancat.model.tok2vec.embed]
@architectures = "spacy.MultiHashEmbed.v1"
width = 96
rows = [5000,2000,1000,1000]
attrs = ["ORTH","PREFIX","SUFFIX","SHAPE"]
include_static_vectors = false

[components.spancat.model.tok2vec.encode]
@architectures = "spacy.MaxoutWindowEncoder.v1"
width = 96
window_size = 1
maxout_pieces = 3
depth = 4

[components.spancat.suggester]
@misc = "spacy.ngram_suggester.v1"
sizes = [1,2,3]

[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 = 0
gold_preproc = false
limit = 0
augmenter = null

[training]
dev_corpus = "corpora.dev"
train_corpus = "corpora.train"
seed = ${system.seed}
gpu_allocator = ${system.gpu_allocator}
dropout = 0.1
accumulate_gradient = 1
patience = 1600
max_epochs = 200
max_steps = 20000
eval_frequency = 200
frozen_components = []
annotating_components = []
before_to_disk = null

[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]
spans_sc_f = 1.0
spans_sc_p = 0.0
spans_sc_r = 0.0

[pretraining]

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

[initialize.components]

[initialize.tokenizer]

[thomashacker]

Hello,
The Spancat component saves the predicted spans to a SpanGroup (in your case spans_key = "sc"). You can access the spans from the SpanGroup by:
docs.spans["sc"]
Then you can access the predicted labels from the span by:
span.label_

[smisra1]

Got it! Thank you! Also, where could I get the score for this predicted span?