Upload tokenizer

README.md

@@ -1,499 +1,498 @@
 ---
-license: apache-2.0
 datasets:
 - lmms-lab/LLaVA-OneVision-Data
 language:
 - en
 - zh
+library_name: transformers
+license: apache-2.0
 metrics:
 - accuracy
-library_name: transformers
 tags:
 - multimodal
-
 model-index:
 - name: llava-onevision-qwen-0.5b-ov
   results:
   - task:
       type: multimodal
     dataset:
-      type: ai2d
       name: AI2D
+      type: ai2d
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 57.1
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: chartqa
       name: ChartQA
+      type: chartqa
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 61.4
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: docvqa
       name: DocVQA
+      type: docvqa
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 73.7
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: infovqa
       name: InfoVQA
+      type: infovqa
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 46.3
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mathverse
       name: MathVerse
+      type: mathverse
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 17.9
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mathvista
       name: MathVista
+      type: mathvista
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 34.8
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mmbench
       name: MMBench
+      type: mmbench
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 52.1
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mme-perception
       name: MME-Perception
+      type: mme-perception
     metrics:
-    - name: score
-      type: score
+    - type: score
       value: 1238
+      name: score
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mme-cognition
       name: MME-Cognition
+      type: mme-cognition
     metrics:
-    - name: score
-      type: score
+    - type: score
       value: 240
-      verified: true      
+      name: score
+      verified: true
   - task:
       type: multimodal
     dataset:
-      type: mmmu
       name: MMMU
+      type: mmmu
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 31.4
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mmvet
       name: MMVet
+      type: mmvet
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 29.1
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mmstar
       name: MMStar
+      type: mmstar
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 37.5
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: seed-bench
       name: Seed-Bench
+      type: seed-bench
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 65.5
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: science-qa
       name: Science-QA
+      type: science-qa
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 67.2
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: imagedc
       name: ImageDC
+      type: imagedc
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 83.3
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mmlbench
       name: MMLBench
+      type: mmlbench
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 49.9
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: realworldqa
       name: RealWorldQA
+      type: realworldqa
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 55.6
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: vibe-eval
       name: Vibe-Eval
+      type: vibe-eval
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 33.8
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: llava-w
       name: LLaVA-W
+      type: llava-w
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 74.2
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: l-wilder
       name: L-Wilder
+      type: l-wilder
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 55.0
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: actnet-qa
       name: ActNet-QA
+      type: actnet-qa
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 50.5
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: egoschema
       name: EgoSchema
+      type: egoschema
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 26.8
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mlvu
       name: MLVU
+      type: mlvu
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 50.3
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mvbench
       name: MVBench
+      type: mvbench
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 45.5
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: nextqa
       name: NextQA
+      type: nextqa
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 57.2
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: percepTest
       name: PercepTest
+      type: percepTest
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 49.2
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: seedbench
       name: SeedBench
+      type: seedbench
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 44.2
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: videochatgpt
       name: VideoChatGPT
+      type: videochatgpt
     metrics:
-    - name: score
-      type: score
+    - type: score
       value: 3.12
+      name: score
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: videodc
       name: VideoDC
+      type: videodc
     metrics:
-    - name: score
-      type: score
+    - type: score
       value: 3.55
+      name: score
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: videomme
       name: VideoMME
+      type: videomme
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 44.0
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: iei
       name: Image Edit Instruction
+      type: iei
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 17.1
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mi-vqa
       name: MI-VQA
+      type: mi-vqa
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 48.7
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: nlvr2
       name: NLVR2
+      type: nlvr2
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 63.4
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: puzzle
       name: Puzzle
+      type: puzzle
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 35.4
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: q-bench
       name: Q-Bench
+      type: q-bench
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 48.8
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: spot-diff
       name: Spot-Diff
+      type: spot-diff
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 36.4
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: tr-vqa
       name: TR-VQA
+      type: tr-vqa
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 65.0
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: vst
       name: VST
+      type: vst
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 29.8
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: scannet-chat
       name: ScanNet-Chat
+      type: scannet-chat
     metrics:
-    - name: accuracy
-      type: accuracy
-      value: 60.00
+    - type: accuracy
+      value: 60.0
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: scannet-td
       name: ScanNet-TD
+      type: scannet-td
     metrics:
-    - name: accuracy
-      type: accuracy
-      value: 48.00
+    - type: accuracy
+      value: 48.0
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: scanqa
       name: ScanQA
+      type: scanqa
     metrics:
-    - name: accuracy
-      type: accuracy
-      value: 29.40
+    - type: accuracy
+      value: 29.4
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: alfred
       name: ALFRED
+      type: alfred
     metrics:
-    - name: accuracy
-      type: accuracy
-      value: 62.20
+    - type: accuracy
+      value: 62.2
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: nuscenesvqa
       name: nuScenesVQA
+      type: nuscenesvqa
     metrics:
-    - name: accuracy
-      type: accuracy
-      value: 70.50
+    - type: accuracy
+      value: 70.5
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: blink
       name: BLINK
+      type: blink
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 52.1
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mantis
       name: Mantis
+      type: mantis
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 39.6
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: mathverse-mv
       name: MathVerse-mv
+      type: mathverse-mv
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 60.0
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: muirbench
       name: MuirBench
+      type: muirbench
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 25.5
+      name: accuracy
       verified: true
   - task:
       type: multimodal
     dataset:
-      type: sciverse-mv
       name: SciVerse-mv
+      type: sciverse-mv
     metrics:
-    - name: accuracy
-      type: accuracy
+    - type: accuracy
       value: 29.1
-      verified: true      
+      name: accuracy
+      verified: true
 ---
 
 

added_tokens.json

@@ -1,4 +1,5 @@
 {
+  "<image>": 151646,
   "<|endoftext|>": 151643,
   "<|im_end|>": 151645,
   "<|im_start|>": 151644

tokenizer.json

@@ -29,6 +29,15 @@
       "rstrip": false,
       "normalized": false,
       "special": true
+    },
+    {
+      "id": 151646,
+      "content": "<image>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
     }
   ],
   "normalizer": {
@@ -73,6 +82,7 @@
     "end_of_word_suffix": "",
     "fuse_unk": false,
     "byte_fallback": false,
+    "ignore_merges": false,
     "vocab": {
       "!": 0,
       "\"": 1,

tokenizer_config.json

@@ -24,6 +24,14 @@
       "rstrip": false,
       "single_word": false,
       "special": true
+    },
+    "151646": {
+      "content": "<image>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
     }
   },
   "additional_special_tokens": [