Update spaCy pipeline

README.md

@@ -8,24 +8,14 @@ model-index:
 - name: en_grammar_checker
   results: []
 ---
-Essay Grammar Checker
-
 | Feature | Description |
 | --- | --- |
 | **Name** | `en_grammar_checker` |
-| **Version** | `1.0.1` |
-| **spaCy** | `>=3.4.4,<3.5.0` |
-| **Default Pipeline** | `errors` |
-| **Components** | `errors` |
+| **Version** | `1.0.0` |
+| **spaCy** | `>=3.3.0,<3.4.0` |
+| **Default Pipeline** | `punctuation`, `spelling`, `articles`, `grammar_major`, `grammar_minor`, `vocabulary` |
+| **Components** | `punctuation`, `spelling`, `articles`, `grammar_major`, `grammar_minor`, `vocabulary` |
 | **Vectors** | 0 keys, 0 unique vectors (0 dimensions) |
 | **Sources** | n/a |
-| **License** | `CC BY-SA 3.0` |
-| **Author** | [Irina Proskurina]() |
-
-### Accuracy
-
-| Type | Score |
-| --- | --- |
-| `SPANS_ERRORS_P` | 79.38 |
-| `SPANS_ERRORS_R` | 44.77 |
-| `SPANS_ERRORS_F` | 57.25 |
+| **License** | n/a |
+| **Author** | [n/a]() |

config.cfg

@@ -1,124 +1,92 @@
 [paths]
-train = "./realec/train.spacy"
-dev = "./realec/dev.spacy"
+train = null
+dev = null
 vectors = null
 init_tok2vec = null
 
 [system]
-gpu_allocator = "pytorch"
 seed = 0
+gpu_allocator = null
 
 [nlp]
 lang = "en"
-pipeline = ["transformer","spancat"]
-batch_size = 16
+pipeline = ["punctuation","spelling","articles","grammar_major","grammar_minor","vocabulary"]
 disabled = []
 before_creation = null
 after_creation = null
 after_pipeline_creation = null
+batch_size = 1000
 tokenizer = {"@tokenizers":"spacy.Tokenizer.v1"}
 
 [components]
 
-[components.spancat]
-factory = "spancat"
-max_positive = null
-scorer = {"@scorers":"spacy.spancat_scorer.v1"}
-spans_key = "errors"
-threshold = 0.5
+[components.articles]
+factory = "articles"
 
-[components.spancat.model]
-@architectures = "spacy.SpanCategorizer.v1"
+[components.grammar_major]
+factory = "grammar_major"
 
-[components.spancat.model.reducer]
-@layers = "spacy.mean_max_reducer.v1"
-hidden_size = 128
+[components.grammar_minor]
+factory = "grammar_minor"
 
-[components.spancat.model.scorer]
-@layers = "spacy.LinearLogistic.v1"
-nO = null
-nI = null
+[components.punctuation]
+factory = "punctuation"
 
-[components.spancat.model.tok2vec]
-@architectures = "spacy-transformers.TransformerListener.v1"
-grad_factor = 1.0
-pooling = {"@layers":"reduce_mean.v1"}
-upstream = "*"
+[components.spelling]
+factory = "spelling"
 
-[components.spancat.suggester]
-@misc = "spacy.ngram_suggester.v1"
-sizes = [1,2,3]
-
-[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.v3"
-name = "bert-base-cased"
-mixed_precision = false
-
-[components.transformer.model.get_spans]
-@span_getters = "spacy-transformers.strided_spans.v1"
-window = 128
-stride = 96
-
-[components.transformer.model.grad_scaler_config]
-
-[components.transformer.model.tokenizer_config]
-use_fast = true
-
-[components.transformer.model.transformer_config]
+[components.vocabulary]
+factory = "vocabulary"
 
 [corpora]
 
 [corpora.dev]
 @readers = "spacy.Corpus.v1"
-path = "./realec/dev.spacy"
-max_length = 0
+path = ${paths.dev}
 gold_preproc = false
+max_length = 0
 limit = 0
 augmenter = null
 
 [corpora.train]
 @readers = "spacy.Corpus.v1"
-path = "./realec/train.spacy"
-max_length = 0
+path = ${paths.train}
 gold_preproc = false
+max_length = 0
 limit = 0
 augmenter = null
 
 [training]
-accumulate_gradient = 3
-dev_corpus = "corpora.dev"
-train_corpus = "corpora.train"
-seed = 0
-gpu_allocator = "pytorch"
+seed = ${system.seed}
+gpu_allocator = ${system.gpu_allocator}
 dropout = 0.1
+accumulate_gradient = 1
 patience = 1600
 max_epochs = 0
 max_steps = 20000
 eval_frequency = 200
 frozen_components = []
 annotating_components = []
+dev_corpus = "corpora.dev"
+train_corpus = "corpora.train"
 before_to_disk = null
 
 [training.batcher]
-@batchers = "spacy.batch_by_padded.v1"
-discard_oversize = true
-size = 2000
-buffer = 256
+@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.WandbLogger.v3"
-project_name = "my-awesome-project"
-remove_config_values = ["paths.train","paths.dev","corpora.train.path","corpora.dev.path"]
-log_dataset_dir = null
-entity = null
-run_name = "grammar-checker"
-model_log_interval = null
+@loggers = "spacy.ConsoleLogger.v1"
+progress_bar = false
 
 [training.optimizer]
 @optimizers = "Adam.v1"
@@ -129,34 +97,15 @@ 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
+learn_rate = 0.001
 
 [training.score_weights]
-spans_sc_f = 0.5
-spans_sc_p = 0.0
-spans_sc_r = 0.0
-spans_Agreement_errors_f = 0.06
-spans_Articles_f = 0.03
-spans_Capitalisation_f = 0.05
-spans_Formational_affixes_f = 0.1
-spans_Noun_number_f = 0.04
-spans_Numerals_f = 0.06
-spans_Prepositions_f = 0.05
-spans_Punctuation_f = 0.03
-spans_Spelling_f = 0.02
-spans_Tense_choice_f = 0.03
-spans_lex_item_choice_f = 0.03
 
 [pretraining]
 
 [initialize]
-vectors = null
-init_tok2vec = null
+vectors = ${paths.vectors}
+init_tok2vec = ${paths.init_tok2vec}
 vocab_data = null
 lookups = null
 before_init = null

custom_factories.py

@@ -9,8 +9,33 @@ meta = get_model_meta(model_path)
 data_dir = f"{meta['lang']}_{meta['name']}-{meta['version']}"
 components_path = model_path / data_dir / "training"
 
-@Language.component("errors")
-def errors(doc):
-    nlp_vocabulary = spacy.load(components_path)
-    print("Loaded component")
-    return nlp_vocabulary(doc)
+@Language.component("vocabulary")
+def vocabulary(doc):
+    nlp_vocabulary = spacy.load(components_path/"vocabulary")
+    return nlp_vocabulary(doc)
+
+@Language.component("articles")
+def articles(doc):
+    nlp_articles = spacy.load(components_path/"articles")
+    return nlp_articles(doc)
+
+@Language.component("punctuation")
+def punctuation(doc):
+    nlp_punctuation = spacy.load(components_path/"punctuation")
+    return nlp_punctuation(doc)
+
+@Language.component("spelling")
+def spelling(doc):
+    nlp_spelling = spacy.load(components_path/"spelling")
+    return nlp_spelling(doc)
+
+@Language.component("grammar_major")
+def grammar_major(doc):
+    nlp_grammar_major = spacy.load(components_path/"grammar_major")
+    return nlp_grammar_major(doc)
+
+@Language.component("grammar_minor")
+def grammar_minor(doc):
+    nlp_grammar_minor = spacy.load(components_path/"grammar_minor")
+    return nlp_grammar_minor(doc)
+

en_grammar_checker-any-py3-none-any.whl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bce34b4aa2e04c41a3fada9ada855e4745a8ca7e8858e63b672f569c2255fdc6
-size 406893070
+oid sha256:f0a808972cae10f9b445a2435a70ee3664f8433086d3b3862febe7679cfb321e
+size 2441454414

meta.json

@@ -1,13 +1,13 @@
 {
   "lang":"en",
   "name":"grammar_checker",
-  "version":"1.0.1",
+  "version":"1.0.0",
   "description":"Essay Grammar Checker",
   "author":"Irina Proskurina",
   "email":"",
   "url":"",
   "license":"CC BY-SA 3.0",
-  "spacy_version":">=3.4.4,<3.5.0",
+  "spacy_version":">=3.3.0,<3.4.0",
   "parent_package":"spacy",
   "requirements":[
     "spacy-transformers>=1.0.0,<1.1.0"
@@ -24,77 +24,128 @@
     "name":null
   },
   "pipeline":[
-    "errors"
+    "punctuation",
+    "spelling",
+    "articles",
+    "grammar_major",
+    "grammar_minor",
+    "vocabulary"
   ],
   "components":[
-    "errors"
+    "punctuation",
+    "spelling",
+    "articles",
+    "grammar_major",
+    "grammar_minor",
+    "vocabulary"
   ],
   "labels":{
 
   },
   "performance":{
-    "spans_errors_p":0.7937892339,
-    "spans_errors_r":0.4476503759,
-    "spans_errors_f":0.5724644939,
-    "spans_errors_per_type":{
-      "Numerals":{
-        "p":0.7313328681,
-        "r":0.577092511,
-        "f":0.6451215759
+    "spans_articles_p":0.8274481176,
+    "spans_articles_r":0.8197202306,
+    "spans_articles_f":0.823566046,
+    "spans_punctuation_p":0.8723066455,
+    "spans_punctuation_r":0.6734372264,
+    "spans_punctuation_f":0.7600790514,
+    "spans_spelling_p":0.9037660231,
+    "spans_spelling_r":0.9159588014,
+    "spans_spelling_f":0.9098215644,
+    "spans_spelling_per_type":{
+      "Spelling":{
+        "p":0.9044968253,
+        "r":0.9227970539,
+        "f":0.9135553018
+      },
+      "Capitalisation":{
+        "p":0.8959762655,
+        "r":0.8483146067,
+        "f":0.8714942736
+      }
+    },
+    "spans_vocabulary_p":0.8326700274,
+    "spans_vocabulary_r":0.4697166234,
+    "spans_vocabulary_f":0.6006188,
+    "spans_vocabulary_per_type":{
+      "lex_part_choice":{
+        "p":0.8783783784,
+        "r":0.0741444867,
+        "f":0.1367461431
       },
       "lex_item_choice":{
-        "p":0.7750791975,
-        "r":0.1950571353,
-        "f":0.3116772824
+        "p":0.8219627873,
+        "r":0.5290802348,
+        "f":0.6437755977
       },
-      "Articles":{
-        "p":0.785046729,
-        "r":0.4552258065,
-        "f":0.5762822607
+      "Category_confusion":{
+        "p":0.84472708,
+        "r":0.5153572583,
+        "f":0.6401606426
       },
-      "Punctuation":{
-        "p":0.6955835962,
-        "r":0.2376077586,
-        "f":0.3542168675
+      "Formational_affixes":{
+        "p":0.8892307692,
+        "r":0.5608020699,
+        "f":0.6878222927
+      }
+    },
+    "spans_grammar_minor_p":0.8690835299,
+    "spans_grammar_minor_r":0.7119557892,
+    "spans_grammar_minor_f":0.7827117807,
+    "spans_grammar_minor_per_type":{
+      "Verb_pattern":{
+        "p":0.7399617591,
+        "r":0.2491951062,
+        "f":0.3728323699
       },
-      "Prepositions":{
-        "p":0.8163471241,
-        "r":0.3766294227,
-        "f":0.5154507805
+      "Noun_number":{
+        "p":0.8889171489,
+        "r":0.9032758286,
+        "f":0.8960389694
       },
-      "Formational_affixes":{
-        "p":0.7269700333,
-        "r":0.6031307551,
-        "f":0.6592853548
+      "Word_order":{
+        "p":0.8179384203,
+        "r":0.2189179506,
+        "f":0.3453928773
+      },
+      "Numerals":{
+        "p":0.7062761506,
+        "r":0.602426838,
+        "f":0.6502311248
       },
+      "Determiners":{
+        "p":0.5853658537,
+        "r":0.0231884058,
+        "f":0.0446096654
+      }
+    },
+    "spans_grammar_major_p":0.8694734373,
+    "spans_grammar_major_r":0.6123730074,
+    "spans_grammar_major_f":0.7186197538,
+    "spans_grammar_major_per_type":{
       "Agreement_errors":{
-        "p":0.7909018356,
-        "r":0.5164147994,
-        "f":0.6248423707
+        "p":0.8953090397,
+        "r":0.7739296902,
+        "f":0.8302062914
       },
-      "Capitalisation":{
-        "p":0.8034148593,
-        "r":0.7899274047,
-        "f":0.7966140471
+      "Prepositions":{
+        "p":0.881741712,
+        "r":0.5715202053,
+        "f":0.6935201401
       },
-      "Noun_number":{
-        "p":0.8251445087,
-        "r":0.5558079169,
-        "f":0.6642109345
+      "Redundant_comp":{
+        "p":0.8028656126,
+        "r":0.314191802,
+        "f":0.4516397999
       },
       "Tense_choice":{
-        "p":0.7827648115,
-        "r":0.5369458128,
-        "f":0.6369612856
-      },
-      "Spelling":{
-        "p":0.886746988,
-        "r":0.4357608052,
-        "f":0.5843588726
+        "p":0.86231523,
+        "r":0.7354107649,
+        "f":0.7938231022
       }
     }
   },
-  "speed":2779.5295317788,
+  "speed":5613.7287946692,
   "spacy_git_version":"61dfdd9fb",
   "disabled":[
 

tokenizer

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24ad4ddf9a27837484c7fe2ae8a454167cfc9604ef2e7740a2e74a39cc2c1bc3
-size 76990
+oid sha256:f0f3f5297dc14c5fd94488174149bd87b5f24dff28e6be993eb3ea7a92e53417
+size 78126

training/articles/config.cfg

@@ -0,0 +1,157 @@
+[paths]
+train = "./realec/articles_train.spacy"
+dev = "./realec/articles_dev.spacy"
+vectors = null
+init_tok2vec = null
+
+[system]
+gpu_allocator = "pytorch"
+seed = 0
+
+[nlp]
+lang = "en"
+pipeline = ["transformer","spancat"]
+batch_size = 128
+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 = "articles"
+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-transformers.TransformerListener.v1"
+grad_factor = 1.0
+pooling = {"@layers":"reduce_mean.v1"}
+upstream = "*"
+
+[components.spancat.suggester]
+@misc = "spacy.ngram_suggester.v1"
+sizes = [1,2,3]
+
+[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.v3"
+name = "bert-base-cased"
+mixed_precision = false
+
+[components.transformer.model.get_spans]
+@span_getters = "spacy-transformers.strided_spans.v1"
+window = 128
+stride = 96
+
+[components.transformer.model.grad_scaler_config]
+
+[components.transformer.model.tokenizer_config]
+use_fast = true
+
+[components.transformer.model.transformer_config]
+
+[corpora]
+
+[corpora.dev]
+@readers = "spacy.Corpus.v1"
+path = "./realec/articles_dev.spacy"
+max_length = 0
+gold_preproc = false
+limit = 0
+augmenter = null
+
+[corpora.train]
+@readers = "spacy.Corpus.v1"
+path = "./realec/articles_train.spacy"
+max_length = 0
+gold_preproc = false
+limit = 0
+augmenter = null
+
+[training]
+accumulate_gradient = 3
+dev_corpus = "corpora.dev"
+train_corpus = "corpora.train"
+frozen_components = []
+seed = 0
+gpu_allocator = "pytorch"
+dropout = 0.1
+patience = 1600
+max_epochs = 0
+max_steps = 20000
+eval_frequency = 200
+annotating_components = []
+before_to_disk = null
+
+[training.batcher]
+@batchers = "spacy.batch_by_padded.v1"
+discard_oversize = true
+size = 2000
+buffer = 256
+get_length = null
+
+[training.logger]
+@loggers = "spacy.WandbLogger.v3"
+project_name = "adwiser"
+remove_config_values = ["paths.train","paths.dev","corpora.train.path","corpora.dev.path"]
+model_log_interval = 1000
+log_dataset_dir = null
+entity = null
+run_name = "articles"
+
+[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]
+spans_sc_f = 0.5
+spans_sc_p = 0.0
+spans_sc_r = 0.0
+spans_Articles_f = 0.5
+
+[pretraining]
+
+[initialize]
+vectors = null
+init_tok2vec = null
+vocab_data = null
+lookups = null
+before_init = null
+after_init = null
+
+[initialize.components]
+
+[initialize.tokenizer]

training/articles/meta.json

@@ -0,0 +1,46 @@
+{
+  "lang":"en",
+  "name":"pipeline",
+  "version":"0.0.0",
+  "spacy_version":">=3.3.0,<3.4.0",
+  "description":"",
+  "author":"",
+  "email":"",
+  "url":"",
+  "license":"",
+  "spacy_git_version":"Unknown",
+  "vectors":{
+    "width":0,
+    "vectors":0,
+    "keys":0,
+    "name":null,
+    "mode":"default"
+  },
+  "labels":{
+    "transformer":[
+
+    ],
+    "spancat":[
+      "Articles"
+    ]
+  },
+  "pipeline":[
+    "transformer",
+    "spancat"
+  ],
+  "components":[
+    "transformer",
+    "spancat"
+  ],
+  "disabled":[
+
+  ],
+  "performance":{
+    "spans_sc_f":0.0,
+    "spans_sc_p":0.0,
+    "spans_sc_r":0.0,
+    "spans_Articles_f":0.0,
+    "transformer_loss":100.8329219741,
+    "spancat_loss":2964.9278720927
+  }
+}

training/articles/spancat/cfg

@@ -0,0 +1,8 @@
+{
+  "labels":[
+    "Articles"
+  ],
+  "spans_key":"articles",
+  "threshold":0.5,
+  "max_positive":null
+}

training/articles/spancat/model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5bcd34ab7d5ff9adb0214059514deb98fb9c131f29d0a0a7a4c520fcafd3e875
+size 4723491

training/articles/tokenizer

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0f3f5297dc14c5fd94488174149bd87b5f24dff28e6be993eb3ea7a92e53417
+size 78126

training/articles/transformer/cfg

@@ -0,0 +1,3 @@
+{
+  "max_batch_items":4096
+}

training/articles/transformer/model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2036ce8a5df6924d046d42dd67dc03f670b4d1d25c2d8073de95822165e9a5ed
+size 434208023

training/articles/vocab/key2row

@@ -0,0 +1 @@
+�

training/articles/vocab/lookups.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76be8b528d0075f7aae98d6fa57a6d3c83ae480a8469e668d7b0af968995ac71
+size 1

training/articles/vocab/strings.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6c403d7808ec8eeeafe465538d7df7f45c1202362b29c9c627a7ed8302dc210
+size 287029

training/articles/vocab/vectors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14772b683e726436d5948ad3fff2b43d036ef2ebbe3458aafed6004e05a40706
+size 128

training/articles/vocab/vectors.cfg

@@ -0,0 +1,3 @@
+{
+  "mode":"default"
+}

training/grammar_major/config.cfg

@@ -0,0 +1,160 @@
+[paths]
+train = "./realec/grammar_major_train.spacy"
+dev = "./realec/grammar_major_dev.spacy"
+vectors = null
+init_tok2vec = null
+
+[system]
+gpu_allocator = "pytorch"
+seed = 0
+
+[nlp]
+lang = "en"
+pipeline = ["transformer","spancat"]
+batch_size = 128
+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 = "grammar_major"
+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-transformers.TransformerListener.v1"
+grad_factor = 1.0
+pooling = {"@layers":"reduce_mean.v1"}
+upstream = "*"
+
+[components.spancat.suggester]
+@misc = "spacy.ngram_suggester.v1"
+sizes = [1,2,3]
+
+[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.v3"
+name = "bert-base-cased"
+mixed_precision = false
+
+[components.transformer.model.get_spans]
+@span_getters = "spacy-transformers.strided_spans.v1"
+window = 128
+stride = 96
+
+[components.transformer.model.grad_scaler_config]
+
+[components.transformer.model.tokenizer_config]
+use_fast = true
+
+[components.transformer.model.transformer_config]
+
+[corpora]
+
+[corpora.dev]
+@readers = "spacy.Corpus.v1"
+path = "./realec/grammar_major_dev.spacy"
+max_length = 0
+gold_preproc = false
+limit = 0
+augmenter = null
+
+[corpora.train]
+@readers = "spacy.Corpus.v1"
+path = "./realec/grammar_major_train.spacy"
+max_length = 0
+gold_preproc = false
+limit = 0
+augmenter = null
+
+[training]
+accumulate_gradient = 3
+dev_corpus = "corpora.dev"
+train_corpus = "corpora.train"
+frozen_components = []
+seed = 0
+gpu_allocator = "pytorch"
+dropout = 0.1
+patience = 1600
+max_epochs = 0
+max_steps = 20000
+eval_frequency = 200
+annotating_components = []
+before_to_disk = null
+
+[training.batcher]
+@batchers = "spacy.batch_by_padded.v1"
+discard_oversize = true
+size = 2000
+buffer = 256
+get_length = null
+
+[training.logger]
+@loggers = "spacy.WandbLogger.v3"
+project_name = "adwiser"
+remove_config_values = ["paths.train","paths.dev","corpora.train.path","corpora.dev.path"]
+model_log_interval = 1000
+log_dataset_dir = null
+entity = null
+run_name = "grammar_major"
+
+[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]
+spans_sc_f = 0.5
+spans_sc_p = 0.0
+spans_sc_r = 0.0
+spans_Agreement_errors_f = 0.13
+spans_Prepositions_f = 0.12
+spans_Redundant_comp_f = 0.14
+spans_Tense_choice_f = 0.11
+
+[pretraining]
+
+[initialize]
+vectors = null
+init_tok2vec = null
+vocab_data = null
+lookups = null
+before_init = null
+after_init = null
+
+[initialize.components]
+
+[initialize.tokenizer]

training/grammar_major/meta.json

@@ -0,0 +1,52 @@
+{
+  "lang":"en",
+  "name":"pipeline",
+  "version":"0.0.0",
+  "spacy_version":">=3.3.0,<3.4.0",
+  "description":"",
+  "author":"",
+  "email":"",
+  "url":"",
+  "license":"",
+  "spacy_git_version":"Unknown",
+  "vectors":{
+    "width":0,
+    "vectors":0,
+    "keys":0,
+    "name":null,
+    "mode":"default"
+  },
+  "labels":{
+    "transformer":[
+
+    ],
+    "spancat":[
+      "Agreement_errors",
+      "Prepositions",
+      "Tense_choice",
+      "Redundant_comp"
+    ]
+  },
+  "pipeline":[
+    "transformer",
+    "spancat"
+  ],
+  "components":[
+    "transformer",
+    "spancat"
+  ],
+  "disabled":[
+
+  ],
+  "performance":{
+    "spans_sc_f":0.0,
+    "spans_sc_p":0.0,
+    "spans_sc_r":0.0,
+    "spans_Agreement_errors_f":0.0,
+    "spans_Prepositions_f":0.0,
+    "spans_Redundant_comp_f":0.0,
+    "spans_Tense_choice_f":0.0,
+    "transformer_loss":155.2905472092,
+    "spancat_loss":4539.9674681051
+  }
+}

training/grammar_major/spancat/cfg

@@ -0,0 +1,11 @@
+{
+  "labels":[
+    "Agreement_errors",
+    "Prepositions",
+    "Tense_choice",
+    "Redundant_comp"
+  ],
+  "spans_key":"grammar_major",
+  "threshold":0.5,
+  "max_positive":null
+}

training/grammar_major/spancat/model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e9fbf9b48ca97b5fe06e25c549763cea2b182321819a81a2ea9b61eede20cfa
+size 4725039

training/grammar_major/tokenizer

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0f3f5297dc14c5fd94488174149bd87b5f24dff28e6be993eb3ea7a92e53417
+size 78126

training/grammar_major/transformer/cfg

@@ -0,0 +1,3 @@
+{
+  "max_batch_items":4096
+}

training/grammar_major/transformer/model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:745dee6d568bdaf5acde6d3cb6fd6cccfed39741771b1f36ea1d94968d43b2ff
+size 434208023

training/grammar_major/vocab/key2row

@@ -0,0 +1 @@
+�

training/grammar_major/vocab/lookups.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76be8b528d0075f7aae98d6fa57a6d3c83ae480a8469e668d7b0af968995ac71
+size 1

training/grammar_major/vocab/strings.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ef59bf11f594ff3dbb5802fd852d888513130a338f18d441d0738d16cfd1d9f
+size 173293

training/grammar_major/vocab/vectors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14772b683e726436d5948ad3fff2b43d036ef2ebbe3458aafed6004e05a40706
+size 128

training/grammar_major/vocab/vectors.cfg

@@ -0,0 +1,3 @@
+{
+  "mode":"default"
+}

training/grammar_minor/config.cfg

@@ -0,0 +1,161 @@
+[paths]
+train = "./realec/grammar_minor_train.spacy"
+dev = "./realec/grammar_minor_dev.spacy"
+vectors = null
+init_tok2vec = null
+
+[system]
+gpu_allocator = "pytorch"
+seed = 0
+
+[nlp]
+lang = "en"
+pipeline = ["transformer","spancat"]
+batch_size = 128
+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 = "grammar_minor"
+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-transformers.TransformerListener.v1"
+grad_factor = 1.0
+pooling = {"@layers":"reduce_mean.v1"}
+upstream = "*"
+
+[components.spancat.suggester]
+@misc = "spacy.ngram_suggester.v1"
+sizes = [1,2,3]
+
+[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.v3"
+name = "bert-base-cased"
+mixed_precision = false
+
+[components.transformer.model.get_spans]
+@span_getters = "spacy-transformers.strided_spans.v1"
+window = 128
+stride = 96
+
+[components.transformer.model.grad_scaler_config]
+
+[components.transformer.model.tokenizer_config]
+use_fast = true
+
+[components.transformer.model.transformer_config]
+
+[corpora]
+
+[corpora.dev]
+@readers = "spacy.Corpus.v1"
+path = "./realec/grammar_minor_dev.spacy"
+max_length = 0
+gold_preproc = false
+limit = 0
+augmenter = null
+
+[corpora.train]
+@readers = "spacy.Corpus.v1"
+path = "./realec/grammar_minor_train.spacy"
+max_length = 0
+gold_preproc = false
+limit = 0
+augmenter = null
+
+[training]
+accumulate_gradient = 3
+dev_corpus = "corpora.dev"
+train_corpus = "corpora.train"
+frozen_components = []
+seed = 0
+gpu_allocator = "pytorch"
+dropout = 0.1
+patience = 1600
+max_epochs = 0
+max_steps = 20000
+eval_frequency = 200
+annotating_components = []
+before_to_disk = null
+
+[training.batcher]
+@batchers = "spacy.batch_by_padded.v1"
+discard_oversize = true
+size = 2000
+buffer = 256
+get_length = null
+
+[training.logger]
+@loggers = "spacy.WandbLogger.v3"
+project_name = "adwiser"
+remove_config_values = ["paths.train","paths.dev","corpora.train.path","corpora.dev.path"]
+model_log_interval = 1000
+log_dataset_dir = null
+entity = null
+run_name = "grammar_minor"
+
+[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]
+spans_sc_f = 0.5
+spans_sc_p = 0.0
+spans_sc_r = 0.0
+spans_Determiners_f = 0.18
+spans_Noun_number_f = 0.01
+spans_Numerals_f = 0.13
+spans_Verb_pattern_f = 0.12
+spans_Word_order_f = 0.07
+
+[pretraining]
+
+[initialize]
+vectors = null
+init_tok2vec = null
+vocab_data = null
+lookups = null
+before_init = null
+after_init = null
+
+[initialize.components]
+
+[initialize.tokenizer]

training/grammar_minor/meta.json

@@ -0,0 +1,54 @@
+{
+  "lang":"en",
+  "name":"pipeline",
+  "version":"0.0.0",
+  "spacy_version":">=3.3.0,<3.4.0",
+  "description":"",
+  "author":"",
+  "email":"",
+  "url":"",
+  "license":"",
+  "spacy_git_version":"Unknown",
+  "vectors":{
+    "width":0,
+    "vectors":0,
+    "keys":0,
+    "name":null,
+    "mode":"default"
+  },
+  "labels":{
+    "transformer":[
+
+    ],
+    "spancat":[
+      "Verb_pattern",
+      "Determiners",
+      "Numerals",
+      "Word_order",
+      "Noun_number"
+    ]
+  },
+  "pipeline":[
+    "transformer",
+    "spancat"
+  ],
+  "components":[
+    "transformer",
+    "spancat"
+  ],
+  "disabled":[
+
+  ],
+  "performance":{
+    "spans_sc_f":0.0,
+    "spans_sc_p":0.0,
+    "spans_sc_r":0.0,
+    "spans_Determiners_f":0.0,
+    "spans_Noun_number_f":0.0,
+    "spans_Numerals_f":0.0,
+    "spans_Verb_pattern_f":0.0,
+    "spans_Word_order_f":0.0,
+    "transformer_loss":74.2849839284,
+    "spancat_loss":3224.212622695
+  }
+}

training/grammar_minor/spancat/cfg

@@ -0,0 +1,12 @@
+{
+  "labels":[
+    "Verb_pattern",
+    "Determiners",
+    "Numerals",
+    "Word_order",
+    "Noun_number"
+  ],
+  "spans_key":"grammar_minor",
+  "threshold":0.5,
+  "max_positive":null
+}

training/grammar_minor/spancat/model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e0283722a11d5a844cd733d76a12574c63e598f946c2d5843a6f81d80ec9dbe
+size 4725555

training/grammar_minor/tokenizer

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0f3f5297dc14c5fd94488174149bd87b5f24dff28e6be993eb3ea7a92e53417
+size 78126

training/grammar_minor/transformer/cfg

@@ -0,0 +1,3 @@
+{
+  "max_batch_items":4096
+}

training/grammar_minor/transformer/model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a214501cb8a98fc0d7934072baa83b46d144b0f255f60afd65c190c2ed876d61
+size 434208023

training/grammar_minor/vocab/key2row

@@ -0,0 +1 @@
+�

training/grammar_minor/vocab/lookups.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76be8b528d0075f7aae98d6fa57a6d3c83ae480a8469e668d7b0af968995ac71
+size 1

training/grammar_minor/vocab/strings.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:169ca15549fd804567ae7d8222e75f320265558fb39b5b90add91046eb7a3723
+size 201573

training/grammar_minor/vocab/vectors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14772b683e726436d5948ad3fff2b43d036ef2ebbe3458aafed6004e05a40706
+size 128

training/grammar_minor/vocab/vectors.cfg

@@ -0,0 +1,3 @@
+{
+  "mode":"default"
+}

training/punctuation/config.cfg

@@ -0,0 +1,157 @@
+[paths]
+train = "./realec/punctuation_train.spacy"
+dev = "./realec/punctuation_dev.spacy"
+vectors = null
+init_tok2vec = null
+
+[system]
+gpu_allocator = "pytorch"
+seed = 0
+
+[nlp]
+lang = "en"
+pipeline = ["transformer","spancat"]
+batch_size = 128
+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 = "punctuation"
+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-transformers.TransformerListener.v1"
+grad_factor = 1.0
+pooling = {"@layers":"reduce_mean.v1"}
+upstream = "*"
+
+[components.spancat.suggester]
+@misc = "spacy.ngram_suggester.v1"
+sizes = [1,2,3]
+
+[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.v3"
+name = "bert-base-cased"
+mixed_precision = false
+
+[components.transformer.model.get_spans]
+@span_getters = "spacy-transformers.strided_spans.v1"
+window = 128
+stride = 96
+
+[components.transformer.model.grad_scaler_config]
+
+[components.transformer.model.tokenizer_config]
+use_fast = true
+
+[components.transformer.model.transformer_config]
+
+[corpora]
+
+[corpora.dev]
+@readers = "spacy.Corpus.v1"
+path = "./realec/punctuation_dev.spacy"
+max_length = 0
+gold_preproc = false
+limit = 0
+augmenter = null
+
+[corpora.train]
+@readers = "spacy.Corpus.v1"
+path = "./realec/punctuation_train.spacy"
+max_length = 0
+gold_preproc = false
+limit = 0
+augmenter = null
+
+[training]
+accumulate_gradient = 3
+dev_corpus = "corpora.dev"
+train_corpus = "corpora.train"
+frozen_components = []
+seed = 0
+gpu_allocator = "pytorch"
+dropout = 0.1
+patience = 1600
+max_epochs = 0
+max_steps = 20000
+eval_frequency = 200
+annotating_components = []
+before_to_disk = null
+
+[training.batcher]
+@batchers = "spacy.batch_by_padded.v1"
+discard_oversize = true
+size = 2000
+buffer = 256
+get_length = null
+
+[training.logger]
+@loggers = "spacy.WandbLogger.v3"
+project_name = "adwiser"
+remove_config_values = ["paths.train","paths.dev","corpora.train.path","corpora.dev.path"]
+model_log_interval = 1000
+log_dataset_dir = null
+entity = null
+run_name = "punctuation"
+
+[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]
+spans_sc_f = 0.5
+spans_sc_p = 0.0
+spans_sc_r = 0.0
+spans_Punctuation_f = 0.5
+
+[pretraining]
+
+[initialize]
+vectors = null
+init_tok2vec = null
+vocab_data = null
+lookups = null
+before_init = null
+after_init = null
+
+[initialize.components]
+
+[initialize.tokenizer]

training/punctuation/meta.json

@@ -0,0 +1,46 @@
+{
+  "lang":"en",
+  "name":"pipeline",
+  "version":"0.0.0",
+  "spacy_version":">=3.3.0,<3.4.0",
+  "description":"",
+  "author":"",
+  "email":"",
+  "url":"",
+  "license":"",
+  "spacy_git_version":"Unknown",
+  "vectors":{
+    "width":0,
+    "vectors":0,
+    "keys":0,
+    "name":null,
+    "mode":"default"
+  },
+  "labels":{
+    "transformer":[
+
+    ],
+    "spancat":[
+      "Punctuation"
+    ]
+  },
+  "pipeline":[
+    "transformer",
+    "spancat"
+  ],
+  "components":[
+    "transformer",
+    "spancat"
+  ],
+  "disabled":[
+
+  ],
+  "performance":{
+    "spans_sc_f":0.0,
+    "spans_sc_p":0.0,
+    "spans_sc_r":0.0,
+    "spans_Punctuation_f":0.0,
+    "transformer_loss":155.9655457946,
+    "spancat_loss":2200.9864988542
+  }
+}

training/punctuation/spancat/cfg

@@ -0,0 +1,8 @@
+{
+  "labels":[
+    "Punctuation"
+  ],
+  "spans_key":"punctuation",
+  "threshold":0.5,
+  "max_positive":null
+}

training/punctuation/spancat/model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80d8d230b00ce3dfdf04de2e52fc4d14b2ac8e98e0c785e889c5e795940c06b2
+size 4723491

training/punctuation/tokenizer

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0f3f5297dc14c5fd94488174149bd87b5f24dff28e6be993eb3ea7a92e53417
+size 78126

training/punctuation/transformer/cfg

@@ -0,0 +1,3 @@
+{
+  "max_batch_items":4096
+}

training/punctuation/transformer/model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a3cab4830ec3d78cf951043924a896eb0146f4647f4adf3160ac035127f6fd6
+size 434208023

training/punctuation/vocab/key2row

@@ -0,0 +1 @@
+�