Training done
Browse files- preprocessor_config.json +4 -4
- tokenizer.json +18 -4
preprocessor_config.json
CHANGED
|
@@ -19,8 +19,8 @@
|
|
| 19 |
"processor_class": "DonutProcessor",
|
| 20 |
"resample": 2,
|
| 21 |
"rescale_factor": 0.00392156862745098,
|
| 22 |
-
"size":
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
}
|
|
|
|
| 19 |
"processor_class": "DonutProcessor",
|
| 20 |
"resample": 2,
|
| 21 |
"rescale_factor": 0.00392156862745098,
|
| 22 |
+
"size": [
|
| 23 |
+
960,
|
| 24 |
+
1280
|
| 25 |
+
]
|
| 26 |
}
|
tokenizer.json
CHANGED
|
@@ -1,7 +1,21 @@
|
|
| 1 |
{
|
| 2 |
"version": "1.0",
|
| 3 |
-
"truncation":
|
| 4 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
"added_tokens": [
|
| 6 |
{
|
| 7 |
"id": 0,
|
|
@@ -567,8 +581,8 @@
|
|
| 567 |
"single_word": false,
|
| 568 |
"lstrip": false,
|
| 569 |
"rstrip": false,
|
| 570 |
-
"normalized":
|
| 571 |
-
"special":
|
| 572 |
}
|
| 573 |
],
|
| 574 |
"normalizer": {
|
|
|
|
| 1 |
{
|
| 2 |
"version": "1.0",
|
| 3 |
+
"truncation": {
|
| 4 |
+
"direction": "Right",
|
| 5 |
+
"max_length": 768,
|
| 6 |
+
"strategy": "LongestFirst",
|
| 7 |
+
"stride": 0
|
| 8 |
+
},
|
| 9 |
+
"padding": {
|
| 10 |
+
"strategy": {
|
| 11 |
+
"Fixed": 768
|
| 12 |
+
},
|
| 13 |
+
"direction": "Right",
|
| 14 |
+
"pad_to_multiple_of": null,
|
| 15 |
+
"pad_id": 1,
|
| 16 |
+
"pad_type_id": 0,
|
| 17 |
+
"pad_token": "<pad>"
|
| 18 |
+
},
|
| 19 |
"added_tokens": [
|
| 20 |
{
|
| 21 |
"id": 0,
|
|
|
|
| 581 |
"single_word": false,
|
| 582 |
"lstrip": false,
|
| 583 |
"rstrip": false,
|
| 584 |
+
"normalized": true,
|
| 585 |
+
"special": false
|
| 586 |
}
|
| 587 |
],
|
| 588 |
"normalizer": {
|