Add DriveLM MLT sample dataset with 309,555 samples

README.md

@@ -0,0 +1,51 @@
+---
+dataset_info:
+  features:
+  - name: scene_id
+    dtype: string
+  - name: frame_id
+    dtype: string
+  - name: qa_id
+    dtype: string
+  - name: scene_description
+    dtype: string
+  - name: category
+    dtype: string
+  - name: question
+    dtype: string
+  - name: answer
+    dtype: string
+  - name: context
+    dtype: string
+  - name: tag
+    list: int32
+  - name: cluster
+    dtype: string
+  - name: layer
+    dtype: string
+  - name: con_up
+    dtype: string
+  - name: con_down
+    dtype: string
+  - name: key_objects
+    dtype: string
+  - name: num_key_objects
+    dtype: int32
+  - name: available_cameras
+    list: string
+  - name: image_paths
+    list: string
+  - name: mlt
+    dtype: string
+  splits:
+  - name: train
+    num_bytes: 675595039.8920109
+    num_examples: 309555
+  download_size: 13514864
+  dataset_size: 675595039.8920109
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+---

data/train-00000-of-00002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4f3f1a0ae2121a6ba6103eb735f42d6bb5abfef4671bc0c33f966fa7b991660
+size 6762493

data/train-00001-of-00002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f77230891c5b2874d7a386bb4aa87a9833fc9baedf33bf0aa033be3cef25ce9
+size 6752371