End of training

README.md

@@ -1,6 +1,6 @@
 ---
 license: mit
-base_model: alexbeta80/donut-test
+base_model: naver-clova-ix/donut-base
 tags:
 - generated_from_trainer
 datasets:
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # donut-test-ddt
 
-This model is a fine-tuned version of [alexbeta80/donut-test](https://huggingface.co/alexbeta80/donut-test) on the imagefolder dataset.
+This model is a fine-tuned version of [naver-clova-ix/donut-base](https://huggingface.co/naver-clova-ix/donut-base) on the imagefolder dataset.
 
 ## Model description
 
@@ -40,7 +40,7 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 10
+- num_epochs: 8
 
 ### Training results
 

added_tokens.json

@@ -1,34 +1,20 @@
 {
   "</s>": 2,
-  "</s_Consignee>": 57536,
-  "</s_Consignee_Country>": 57533,
-  "</s_Consignor>": 57549,
-  "</s_Consignor_Country>": 57535,
-  "</s_Goods_Description>": 57541,
-  "</s_Marks>": 57548,
-  "</s_Notify>": 57540,
-  "</s_Notify_Country>": 57529,
-  "</s_Second_Notify>": 57546,
-  "</s_Second_Notify_Country>": 57532,
-  "</s_Total_Packages>": 57550,
-  "</s_Total_Volume>": 57547,
-  "</s_Total_Weight>": 57544,
+  "</s_Codici>": 57531,
+  "</s_Data>": 57536,
+  "</s_Fornitore>": 57533,
+  "</s_Numero>": 57529,
+  "</s_Posizioni>": 57530,
+  "</s_Quantita>": 57528,
   "<mask>": 57521,
   "<pad>": 1,
   "<s>": 0,
-  "<s_Consignee>": 57545,
-  "<s_Consignee_Country>": 57528,
-  "<s_Consignor>": 57530,
-  "<s_Consignor_Country>": 57527,
-  "<s_Goods_Description>": 57534,
-  "<s_Marks>": 57538,
-  "<s_Notify>": 57525,
-  "<s_Notify_Country>": 57531,
-  "<s_Second_Notify>": 57542,
-  "<s_Second_Notify_Country>": 57537,
-  "<s_Total_Packages>": 57526,
-  "<s_Total_Volume>": 57543,
-  "<s_Total_Weight>": 57539,
+  "<s_Codici>": 57527,
+  "<s_Data>": 57526,
+  "<s_Fornitore>": 57535,
+  "<s_Numero>": 57534,
+  "<s_Posizioni>": 57525,
+  "<s_Quantita>": 57532,
   "<s_iitcdip>": 57523,
   "<s_synthdog>": 57524,
   "<sep/>": 57522,

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "alexbeta80/donut-test",
+  "_name_or_path": "naver-clova-ix/donut-base",
   "architectures": [
     "VisionEncoderDecoderModel"
   ],
@@ -85,7 +85,7 @@
     "typical_p": 1.0,
     "use_bfloat16": false,
     "use_cache": true,
-    "vocab_size": 57551
+    "vocab_size": 57537
   },
   "decoder_start_token_id": 0,
   "encoder": {
@@ -124,8 +124,8 @@
       "1": "LABEL_1"
     },
     "image_size": [
-      1169,
-      827
+      700,
+      520
     ],
     "initializer_range": 0.02,
     "is_decoder": false,

generation_config.json

@@ -1,10 +1,8 @@
 {
   "_from_model_config": true,
   "bos_token_id": 0,
-  "decoder_start_token_id": 0,
   "eos_token_id": 2,
   "forced_eos_token_id": 2,
-  "max_length": 512,
   "pad_token_id": 1,
   "transformers_version": "4.34.0"
 }

preprocessor_config.json

@@ -19,8 +19,8 @@
   "processor_class": "DonutProcessor",
   "resample": 2,
   "rescale_factor": 0.00392156862745098,
-  "size": {
-    "height": 1169,
-    "width": 827
-  }
+  "size": [
+    520,
+    700
+  ]
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:17e96dcdcf9a2aced71e90c4991db40b87bd3e51312835f729dc5ce38c8777e6
-size 809285846
+oid sha256:d5470c283cff8de1c4095140891f380a5e2774fe2278a27b30269dca234488ae
+size 809228502

special_tokens_map.json

@@ -1,38 +1,21 @@
 {
   "additional_special_tokens": [
-    "<s>",
-    "<pad>",
-    "</s>",
-    "<unk>",
-    "<mask>",
     "<s_iitcdip>",
     "<s_synthdog>",
-    "<s_Notify>",
-    "<s_Total_Packages>",
-    "<s_Consignor_Country>",
-    "<s_Consignee_Country>",
-    "</s_Notify_Country>",
-    "<s_Consignor>",
-    "<s_Notify_Country>",
-    "</s_Second_Notify_Country>",
-    "</s_Consignee_Country>",
-    "<s_Goods_Description>",
-    "</s_Consignor_Country>",
-    "</s_Consignee>",
-    "<s_Second_Notify_Country>",
-    "<s_Marks>",
-    "<s_Total_Weight>",
-    "</s_Notify>",
-    "</s_Goods_Description>",
-    "<s_Second_Notify>",
-    "<s_Total_Volume>",
-    "</s_Total_Weight>",
-    "<s_Consignee>",
-    "</s_Second_Notify>",
-    "</s_Total_Volume>",
-    "</s_Marks>",
-    "</s_Consignor>",
-    "</s_Total_Packages>"
+    "<s_Posizioni>",
+    "<s_Data>",
+    "<s_Codici>",
+    "</s>",
+    "</s_Quantita>",
+    "</s_Numero>",
+    "</s_Posizioni>",
+    "</s_Codici>",
+    "<s_Quantita>",
+    "</s_Fornitore>",
+    "<s_Numero>",
+    "<s_Fornitore>",
+    "<s>",
+    "</s_Data>"
   ],
   "bos_token": "<s>",
   "cls_token": "<s>",

tokenizer_config.json

@@ -2,9 +2,9 @@
   "added_tokens_decoder": {
     "0": {
       "content": "<s>",
-      "lstrip": false,
+      "lstrip": true,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
@@ -18,9 +18,9 @@
     },
     "2": {
       "content": "</s>",
-      "lstrip": false,
+      "lstrip": true,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     },
@@ -35,7 +35,7 @@
     "57521": {
       "content": "<mask>",
       "lstrip": true,
-      "normalized": false,
+      "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
@@ -65,7 +65,7 @@
       "special": true
     },
     "57525": {
-      "content": "<s_Notify>",
+      "content": "<s_Posizioni>",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
@@ -73,7 +73,7 @@
       "special": true
     },
     "57526": {
-      "content": "<s_Total_Packages>",
+      "content": "<s_Data>",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
@@ -81,7 +81,7 @@
       "special": true
     },
     "57527": {
-      "content": "<s_Consignor_Country>",
+      "content": "<s_Codici>",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
@@ -89,7 +89,7 @@
       "special": true
     },
     "57528": {
-      "content": "<s_Consignee_Country>",
+      "content": "</s_Quantita>",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
@@ -97,7 +97,7 @@
       "special": true
     },
     "57529": {
-      "content": "</s_Notify_Country>",
+      "content": "</s_Numero>",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
@@ -105,7 +105,7 @@
       "special": true
     },
     "57530": {
-      "content": "<s_Consignor>",
+      "content": "</s_Posizioni>",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
@@ -113,7 +113,7 @@
       "special": true
     },
     "57531": {
-      "content": "<s_Notify_Country>",
+      "content": "</s_Codici>",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
@@ -121,7 +121,7 @@
       "special": true
     },
     "57532": {
-      "content": "</s_Second_Notify_Country>",
+      "content": "<s_Quantita>",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
@@ -129,7 +129,7 @@
       "special": true
     },
     "57533": {
-      "content": "</s_Consignee_Country>",
+      "content": "</s_Fornitore>",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
@@ -137,7 +137,7 @@
       "special": true
     },
     "57534": {
-      "content": "<s_Goods_Description>",
+      "content": "<s_Numero>",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
@@ -145,7 +145,7 @@
       "special": true
     },
     "57535": {
-      "content": "</s_Consignor_Country>",
+      "content": "<s_Fornitore>",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
@@ -153,119 +153,7 @@
       "special": true
     },
     "57536": {
-      "content": "</s_Consignee>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57537": {
-      "content": "<s_Second_Notify_Country>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57538": {
-      "content": "<s_Marks>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57539": {
-      "content": "<s_Total_Weight>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57540": {
-      "content": "</s_Notify>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57541": {
-      "content": "</s_Goods_Description>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57542": {
-      "content": "<s_Second_Notify>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57543": {
-      "content": "<s_Total_Volume>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57544": {
-      "content": "</s_Total_Weight>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57545": {
-      "content": "<s_Consignee>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57546": {
-      "content": "</s_Second_Notify>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57547": {
-      "content": "</s_Total_Volume>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57548": {
-      "content": "</s_Marks>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57549": {
-      "content": "</s_Consignor>",
-      "lstrip": true,
-      "normalized": false,
-      "rstrip": true,
-      "single_word": false,
-      "special": true
-    },
-    "57550": {
-      "content": "</s_Total_Packages>",
+      "content": "</s_Data>",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
@@ -274,57 +162,33 @@
     }
   },
   "additional_special_tokens": [
-    "<s>",
-    "<pad>",
-    "</s>",
-    "<unk>",
-    "<mask>",
     "<s_iitcdip>",
     "<s_synthdog>",
-    "<s_Notify>",
-    "<s_Total_Packages>",
-    "<s_Consignor_Country>",
-    "<s_Consignee_Country>",
-    "</s_Notify_Country>",
-    "<s_Consignor>",
-    "<s_Notify_Country>",
-    "</s_Second_Notify_Country>",
-    "</s_Consignee_Country>",
-    "<s_Goods_Description>",
-    "</s_Consignor_Country>",
-    "</s_Consignee>",
-    "<s_Second_Notify_Country>",
-    "<s_Marks>",
-    "<s_Total_Weight>",
-    "</s_Notify>",
-    "</s_Goods_Description>",
-    "<s_Second_Notify>",
-    "<s_Total_Volume>",
-    "</s_Total_Weight>",
-    "<s_Consignee>",
-    "</s_Second_Notify>",
-    "</s_Total_Volume>",
-    "</s_Marks>",
-    "</s_Consignor>",
-    "</s_Total_Packages>"
+    "<s_Posizioni>",
+    "<s_Data>",
+    "<s_Codici>",
+    "</s>",
+    "</s_Quantita>",
+    "</s_Numero>",
+    "</s_Posizioni>",
+    "</s_Codici>",
+    "<s_Quantita>",
+    "</s_Fornitore>",
+    "<s_Numero>",
+    "<s_Fornitore>",
+    "<s>",
+    "</s_Data>"
   ],
   "bos_token": "<s>",
   "clean_up_tokenization_spaces": true,
   "cls_token": "<s>",
   "eos_token": "</s>",
   "mask_token": "<mask>",
-  "max_length": 512,
   "model_max_length": 1000000000000000019884624838656,
-  "pad_to_multiple_of": null,
   "pad_token": "<pad>",
-  "pad_token_type_id": 0,
-  "padding_side": "right",
   "processor_class": "DonutProcessor",
   "sep_token": "</s>",
   "sp_model_kwargs": {},
-  "stride": 0,
   "tokenizer_class": "XLMRobertaTokenizer",
-  "truncation_side": "right",
-  "truncation_strategy": "longest_first",
   "unk_token": "<unk>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c5d2c1f955567456d8485e5d41d98208e55524d44111b76444f3a7eead18c68
+oid sha256:2ed308013c1c7b71d331e7af16d7ef386cd223ba7455751efed12a5f11afd8a5
 size 4664