initial commit

added_tokens.json

@@ -0,0 +1 @@
+{"<sep/>": 57522, "<s_iitcdip>": 57523, "<s_synthdog>": 57524, "</s_menu>": 57525, "<s_menu>": 57526, "</s_nm>": 57527, "<s_nm>": 57528, "</s_cnt>": 57529, "<s_cnt>": 57530, "</s_price>": 57531, "<s_price>": 57532, "</s_sub_total>": 57533, "<s_sub_total>": 57534, "</s_subtotal_price>": 57535, "<s_subtotal_price>": 57536, "</s_service_price>": 57537, "<s_service_price>": 57538, "</s_tax_price>": 57539, "<s_tax_price>": 57540, "</s_etc>": 57541, "<s_etc>": 57542, "</s_total>": 57543, "<s_total>": 57544, "</s_total_price>": 57545, "<s_total_price>": 57546, "</s_sub_nm>": 57547, "<s_sub_nm>": 57548, "</s_cashprice>": 57549, "<s_cashprice>": 57550, "</s_changeprice>": 57551, "<s_changeprice>": 57552, "</s_menutype_cnt>": 57553, "<s_menutype_cnt>": 57554, "</s_menuqty_cnt>": 57555, "<s_menuqty_cnt>": 57556, "</s_discount_price>": 57557, "<s_discount_price>": 57558, "</s_unitprice>": 57559, "<s_unitprice>": 57560, "</s_total_etc>": 57561, "<s_total_etc>": 57562, "</s_creditcardprice>": 57563, "<s_creditcardprice>": 57564, "</s_num>": 57565, "<s_num>": 57566, "</s_sub_cnt>": 57567, "<s_sub_cnt>": 57568, "</s_discountprice>": 57569, "<s_discountprice>": 57570, "</s_sub_price>": 57571, "<s_sub_price>": 57572, "</s_emoneyprice>": 57573, "<s_emoneyprice>": 57574, "</s_sub_unitprice>": 57575, "<s_sub_unitprice>": 57576, "</s_sub_etc>": 57577, "<s_sub_etc>": 57578, "</s_void_menu>": 57579, "<s_void_menu>": 57580, "</s_othersvc_price>": 57581, "<s_othersvc_price>": 57582, "</s_vatyn>": 57583, "<s_vatyn>": 57584, "</s_itemsubtotal>": 57585, "<s_itemsubtotal>": 57586, "<s_cord>": 57587}

config.json

@@ -0,0 +1,24 @@
+{
+  "_name_or_path": "naver-clova-ix/donut-base",
+  "align_long_axis": false,
+  "architectures": [
+    "DonutModel"
+  ],
+  "decoder_layer": 4,
+  "encoder_layer": [
+    2,
+    2,
+    14,
+    2
+  ],
+  "input_size": [
+    1280,
+    960
+  ],
+  "max_length": 768,
+  "max_position_embeddings": 768,
+  "model_type": "donut",
+  "torch_dtype": "float32",
+  "transformers_version": "4.11.3",
+  "window_size": 10
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6d20983fbce7d629183b320bd00fa5953a816cb867759106049d523d670677c
+size 858616323

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb9e3dce4c326195d08fc3dd0f7e2eee1da8595c847bf4c1a9c78b7a82d47e2d
+size 1296245

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true}, "additional_special_tokens": ["<s_cord>"]}

tokenizer_config.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "sp_model_kwargs": {}, "special_tokens_map_file": null, "tokenizer_file": "/root/.cache/huggingface/transformers/213c2041358e63047b407f94cde1ae23904d31a3bceb57eab291028c1e949437.7135a4b25ac726e19641f0d68803ff02bad960d6319064f55fa9c536929b86fc", "name_or_path": "/mnt/ocr-nfsx1/donut/pretrained_models_hf_format/donut-base", "tokenizer_class": "XLMRobertaTokenizer"}