Upload processor

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,8 +1,8 @@
 ---
-license: apache-2.0
 language:
 - en
 library_name: transformers
+license: apache-2.0
 pipeline_tag: image-text-to-text
 tags:
 - multimodal

added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<pad>": 100352
+}

preprocessor_config.json

@@ -1,15 +1,17 @@
 {
   "_transform": null,
   "auto_map": {
-    "AutoImageProcessor": "vision_processor.AriaVisionProcessor",
-    "AutoProcessor": "processing_aria.AriaProcessor"
+    "AutoImageProcessor": "rhymes-ai/Aria--vision_processor.AriaVisionProcessor",
+    "AutoProcessor": "rhymes-ai/Aria--processing_aria.AriaProcessor"
   },
+  "do_convert_rgb": true,
+  "do_normalize": true,
   "image_mean": [
     0.5,
     0.5,
     0.5
   ],
-  "image_processor_type": "AriaVisionProcessor",
+  "image_processor_type": "AriaImageProcessor",
   "image_std": [
     0.5,
     0.5,
@@ -17,5 +19,85 @@
   ],
   "max_image_size": 980,
   "min_image_size": 336,
-  "processor_class": "AriaProcessor"
+  "processor_class": "AriaProcessor",
+  "resample": 3,
+  "split_image": false,
+  "split_resolutions": [
+    [
+      490,
+      980
+    ],
+    [
+      490,
+      1470
+    ],
+    [
+      490,
+      1960
+    ],
+    [
+      490,
+      2450
+    ],
+    [
+      490,
+      2940
+    ],
+    [
+      490,
+      3430
+    ],
+    [
+      490,
+      3920
+    ],
+    [
+      980,
+      1960
+    ],
+    [
+      980,
+      1470
+    ],
+    [
+      980,
+      980
+    ],
+    [
+      980,
+      490
+    ],
+    [
+      1470,
+      490
+    ],
+    [
+      1470,
+      980
+    ],
+    [
+      1960,
+      490
+    ],
+    [
+      1960,
+      980
+    ],
+    [
+      2450,
+      490
+    ],
+    [
+      2940,
+      490
+    ],
+    [
+      3430,
+      490
+    ],
+    [
+      3920,
+      490
+    ]
+  ]
 }

processor_config.json

@@ -0,0 +1,7 @@
+{
+  "processor_class": "AriaProcessor",
+  "size_conversion": {
+    "490": 128,
+    "980": 256
+  }
+}

special_tokens_map.json

@@ -1,4 +1,12 @@
 {
+  "image_token": "<|img|>",
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
   "unk_token": {
     "content": "<unk>",
     "lstrip": false,

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02702cce6c4de786d52a3dc624b39e86134c159b7490ea30630739c6f723e7f8
+size 11091481