README.md

---
dataset_info:
  features:
  - name: id
    dtype: int64
  - name: session
    dtype: string
  - name: timestamp
    dtype: int64
  - name: word
    dtype: string
  - name: canvas_width
    dtype: float64
  - name: canvas_height
    dtype: float64
  - name: orientation
    dtype: string
  - name: data
    list:
    - name: t
      dtype: int64
    - name: x
      dtype: float64
    - name: y
      dtype: float64
  - name: sentence
    dtype: string
  - name: word_idx
    dtype: int64
  - name: potentially_invalid_sentence
    dtype: bool
  - name: distance
    dtype: float64
  - name: masked_sentence
    dtype: string
  - name: trajectory_sampled
    sequence:
      sequence: float64
  - name: trajectory_sampled_keys
    sequence: string
  - name: trajectory_word
    dtype: string
  splits:
  - name: train
    num_bytes: 27795316758
    num_examples: 9395500
  - name: validation
    num_bytes: 1627250297
    num_examples: 542690
  - name: test
    num_bytes: 1461745596
    num_examples: 499700
  download_size: 9383480056
  dataset_size: 30884312651
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---