initial release

README.md

@@ -0,0 +1,28 @@
+---
+language:
+- "th"
+tags:
+- "thai"
+- "masked-lm"
+- "wikipedia"
+license: "apache-2.0"
+pipeline_tag: "fill-mask"
+mask_token: "<mask>"
+widget:
+- text: "แผนกนี้กำลังเผชิญกับความ<mask>ใหม่"
+---
+
+# roberta-base-thai-syllable
+
+## Model Description
+
+This is a RoBERTa model pre-trained on Thai Wikipedia texts, derived from [wangchanberta-base-wiki-syllable](https://huggingface.co/airesearch/wangchanberta-base-wiki-syllable). Character-embeddings are modified to use BertTokenizerFast. You can fine-tune `roberta-base-thai-syllable` for downstream tasks, such as POS-tagging, dependency-parsing, and so on.
+
+## How to Use
+
+```py
+from transformers import AutoTokenizer,AutoModelForMaskedLM
+tokenizer=AutoTokenizer.from_pretrained("KoichiYasuoka/roberta-base-thai-syllable")
+model=AutoModelForMaskedLM.from_pretrained("KoichiYasuoka/roberta-base-thai-syllable")
+```
+

config.json

@@ -0,0 +1,30 @@
+{
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "mask_token_id": 4,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "tokenizer_class": "BertTokenizerFast",
+  "torch_dtype": "float32",
+  "transformers_version": "4.10.0",
+  "type_vocab_size": 1,
+  "unk_token_id": 3,
+  "use_cache": true,
+  "vocab_size": 33144
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a586df08fb40079c21f555426486f93c9a7ff6e82d75cf095b0c83894a5a7e9
+size 446197851

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": false, "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>", "tokenize_chinese_chars": true, "strip_accents": false, "model_max_length": 416, "do_basic_tokenize": true, "never_split": null, "tokenizer_class": "BertTokenizer"}