DDTdonut-test-tmp

README.md

@@ -16,8 +16,6 @@ should probably proofread and complete it, then remove this comment. -->
 # DDTdonut-test-tmp
 
 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.
-It achieves the following results on the evaluation set:
-- Loss: 0.5462
 
 ## Model description
 
@@ -45,15 +43,6 @@ The following hyperparameters were used during training:
 - num_epochs: 3
 - mixed_precision_training: Native AMP
 
-### Training results
-
-| Training Loss | Epoch | Step | Validation Loss |
-|:-------------:|:-----:|:----:|:---------------:|
-| 3.3689        | 1.0   | 261  | 1.0382          |
-| 0.9267        | 2.0   | 522  | 0.6386          |
-| 0.6698        | 3.0   | 783  | 0.5462          |
-
-
 ### Framework versions
 
 - Transformers 4.38.2

added_tokens.json

@@ -1,18 +1,4 @@
 {
-  "</s_CODICI>": 57534,
-  "</s_DATA>": 57528,
-  "</s_FORNITORE>": 57526,
-  "</s_NUMERO>": 57530,
-  "</s_OGGETTI>": 57532,
-  "</s_POSIZIONI>": 57538,
-  "</s_QUANTITA>": 57536,
-  "<s_CODICI>": 57533,
-  "<s_DATA>": 57527,
-  "<s_FORNITORE>": 57525,
-  "<s_NUMERO>": 57529,
-  "<s_OGGETTI>": 57531,
-  "<s_POSIZIONI>": 57537,
-  "<s_QUANTITA>": 57535,
   "<s_iitcdip>": 57523,
   "<s_synthdog>": 57524,
   "<sep/>": 57522

config.json

@@ -85,7 +85,7 @@
     "typical_p": 1.0,
     "use_bfloat16": false,
     "use_cache": true,
-    "vocab_size": 57539
+    "vocab_size": 57525
   },
   "decoder_start_token_id": 0,
   "encoder": {

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a46d982568a8567b62c7c5336d0dd0d1f2580693312525eb0dbe537a600939aa
-size 809128088
+oid sha256:a489f5f1204286191cd38cfebacaeb876a5f9da99da6a1b4a633bf65508c39b9
+size 809070744

special_tokens_map.json

@@ -1,103 +1,5 @@
 {
   "additional_special_tokens": [
-    {
-      "content": "<s_FORNITORE>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "</s_FORNITORE>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "<s_DATA>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "</s_DATA>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "<s_NUMERO>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "</s_NUMERO>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "<s_OGGETTI>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "</s_OGGETTI>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "<s_CODICI>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "</s_CODICI>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "<s_QUANTITA>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "</s_QUANTITA>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "<s_POSIZIONI>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "</s_POSIZIONI>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
     {
       "content": "<s>",
       "lstrip": false,

tokenizer.json

@@ -1,21 +1,7 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
-  "padding": {
-    "strategy": {
-      "Fixed": 512
-    },
-    "direction": "Right",
-    "pad_to_multiple_of": null,
-    "pad_id": 1,
-    "pad_type_id": 0,
-    "pad_token": "<pad>"
-  },
+  "truncation": null,
+  "padding": null,
   "added_tokens": [
     {
       "id": 0,
@@ -88,132 +74,6 @@
       "rstrip": false,
       "normalized": false,
       "special": true
-    },
-    {
-      "id": 57525,
-      "content": "<s_FORNITORE>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57526,
-      "content": "</s_FORNITORE>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57527,
-      "content": "<s_DATA>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57528,
-      "content": "</s_DATA>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57529,
-      "content": "<s_NUMERO>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57530,
-      "content": "</s_NUMERO>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57531,
-      "content": "<s_OGGETTI>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57532,
-      "content": "</s_OGGETTI>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57533,
-      "content": "<s_CODICI>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57534,
-      "content": "</s_CODICI>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57535,
-      "content": "<s_QUANTITA>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57536,
-      "content": "</s_QUANTITA>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57537,
-      "content": "<s_POSIZIONI>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 57538,
-      "content": "</s_POSIZIONI>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
     }
   ],
   "normalizer": {

tokenizer_config.json

@@ -63,135 +63,9 @@
       "rstrip": false,
       "single_word": false,
       "special": true
-    },
-    "57525": {
-      "content": "<s_FORNITORE>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57526": {
-      "content": "</s_FORNITORE>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57527": {
-      "content": "<s_DATA>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57528": {
-      "content": "</s_DATA>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57529": {
-      "content": "<s_NUMERO>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57530": {
-      "content": "</s_NUMERO>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57531": {
-      "content": "<s_OGGETTI>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57532": {
-      "content": "</s_OGGETTI>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57533": {
-      "content": "<s_CODICI>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57534": {
-      "content": "</s_CODICI>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57535": {
-      "content": "<s_QUANTITA>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57536": {
-      "content": "</s_QUANTITA>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57537": {
-      "content": "<s_POSIZIONI>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "57538": {
-      "content": "</s_POSIZIONI>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
     }
   },
   "additional_special_tokens": [
-    "<s_FORNITORE>",
-    "</s_FORNITORE>",
-    "<s_DATA>",
-    "</s_DATA>",
-    "<s_NUMERO>",
-    "</s_NUMERO>",
-    "<s_OGGETTI>",
-    "</s_OGGETTI>",
-    "<s_CODICI>",
-    "</s_CODICI>",
-    "<s_QUANTITA>",
-    "</s_QUANTITA>",
-    "<s_POSIZIONI>",
-    "</s_POSIZIONI>",
     "<s>",
     "</s>"
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ef50a0eee38a1dbd2b49206427532166bf035d0d8840fab4edff84b4992f2579
+oid sha256:ecdc6dbe548cfe1c2f8f57b396e8bcf49fc1da3cc0360d160fc02952a8680b47
 size 5048