initial release

README.md

@@ -1,3 +1,27 @@
----
-license: cc-by-sa-4.0
----
+---
+language:
+- "be"
+tags:
+- "belarusian"
+- "masked-lm"
+license: "cc-by-sa-4.0"
+datasets:
+- "cc100"
+pipeline_tag: "fill-mask"
+mask_token: "[MASK]"
+---
+
+# roberta-small-belarusian
+
+## Model Description
+
+This is a RoBERTa model pre-trained on [CC-100](https://data.statmt.org/cc-100/). You can fine-tune `roberta-small-belarusian` for downstream tasks, such as [POS-tagging](https://huggingface.co/KoichiYasuoka/roberta-small-belarusian-upos), dependency-parsing, and so on.
+
+## How to Use
+
+```py
+from transformers import AutoTokenizer,AutoModelForMaskedLM
+tokenizer=AutoTokenizer.from_pretrained("KoichiYasuoka/roberta-small-belarusian")
+model=AutoModelForMaskedLM.from_pretrained("KoichiYasuoka/roberta-small-belarusian")
+```
+

config.json

@@ -0,0 +1,27 @@
+{
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 2,
+  "classifier_dropout": null,
+  "eos_token_id": 3,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 256,
+  "initializer_range": 0.02,
+  "intermediate_size": 768,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 128,
+  "model_type": "roberta",
+  "num_attention_heads": 4,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "tokenizer_class": "BertTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.14.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d43cced63354e971498614f2261a3e35557225625931b074194627bba5357827
+size 62927659

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": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": false, "do_lowercase": false, "never_split": ["[CLS]", "[PAD]", "[SEP]", "[UNK]", "[MASK]"], "do_basic_tokenize": true, "model_max_length": 128, "tokenizer_class": "BertTokenizer"}