Add ner configs

README.md

@@ -0,0 +1,33 @@
+---
+language: en
+license: mit
+tags:
+  - ner
+widget:
+  - text: "These shoes from Tommy Hilfiger fit quite well sincerely, Oscar"
+---
+
+### Description
+
+A Named Entity Recognition model trained on a customer feedback data using DistilBert.
+Possible labels are in BIO-notation:
+
+- PROD: for certain products
+- BRND: for brands
+- PERS: people names
+
+The following tags are simply in place to help better categorize the previous tags
+
+- MATR: relating to materials, e.g. cloth, leather, seam, etc.
+- TIME: time related entities
+- MISC: any other entities that might skew ther results
+
+### Usage
+
+```
+from transformers import AutoTokenizer, AutoModelForTokenClassification
+
+  tokenizer = AutoTokenizer.from_pretrained("CouchCat/ma_ner_v7_distil")
+
+  model = AutoModelForTokenClassification.from_pretrained("CouchCat/ma_ner_v7_distil")
+```

config.json

@@ -0,0 +1,48 @@
+{
+  "_name_or_path": "distilbert-base-cased",
+  "activation": "gelu",
+  "architectures": [
+    "DistilBertForTokenClassification"
+  ],
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "hidden_dim": 3072,
+  "id2label": {
+    "0": "B-BRND",
+    "1": "B-MATR",
+    "2": "B-MISC",
+    "3": "B-PERS",
+    "4": "B-PROD",
+    "5": "B-TIME",
+    "6": "I-BRND",
+    "7": "I-PERS",
+    "8": "O",
+    "9": "PAD"
+  },
+  "initializer_range": 0.02,
+  "label2id": {
+    "B-BRND": 0,
+    "B-MATR": 1,
+    "B-MISC": 2,
+    "B-PERS": 3,
+    "B-PROD": 4,
+    "B-TIME": 5,
+    "I-BRND": 6,
+    "I-PERS": 7,
+    "O": 8,
+    "PAD": 9
+  },
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "output_past": true,
+  "pad_token_id": 0,
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "tie_weights_": true,
+  "transformers_version": "4.3.2",
+  "vocab_size": 28996
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96ea11f5b2435ac079502bf2080ff6eef7d69b4e9d2f35a94eddb7c2c5b02f12
+size 260835627

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": false, "do_basic_tokenize": true, "never_split": null, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "name_or_path": "distilbert-base-cased"}