id string | image string | conversations list |
|---|---|---|
coding_session_0 | coding_session/frame_000000.jpg | [
{
"from": "human",
"value": "<image>\nWhat is happening in this scene?"
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: handle; tool: n/a; attention: away]"
}
] |
coding_session_1 | coding_session/frame_000001.jpg | [
{
"from": "human",
"value": "<image>\nDescribe what the person is doing."
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: handle; tool: n/a; attention: away]"
}
] |
coding_session_2 | coding_session/frame_000002.jpg | [
{
"from": "human",
"value": "<image>\nWhat activity and tools are visible here?"
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: handle; tool: n/a; attention: away]"
}
] |
coding_session_3 | coding_session/frame_000003.jpg | [
{
"from": "human",
"value": "<image>\nWhat is happening in this scene?"
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: handle; tool: n/a; attention: away]"
}
] |
coding_session_4 | coding_session/frame_000004.jpg | [
{
"from": "human",
"value": "<image>\nDescribe what the person is doing."
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: handle; tool: n/a; attention: away]"
}
] |
coding_session_5 | coding_session/frame_000005.jpg | [
{
"from": "human",
"value": "<image>\nWhat activity and tools are visible here?"
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: handle; tool: n/a; attention: away]"
}
] |
coding_session_6 | coding_session/frame_000006.jpg | [
{
"from": "human",
"value": "<image>\nWhat is happening in this scene?"
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: idle; tool: n/a; attention: away]"
}
] |
coding_session_7 | coding_session/frame_000007.jpg | [
{
"from": "human",
"value": "<image>\nDescribe what the person is doing."
},
{
"from": "gpt",
"value": "C objects: bottle, pen; activity: leave_desk; reading a page\n[verb: leave; tool: n/a; attention: away]"
}
] |
reading_session_0 | reading_session/frame_000000.jpg | [
{
"from": "human",
"value": "<image>\nWhat is happening in this scene?"
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: idle; tool: n/a; attention: away]"
}
] |
reading_session_1 | reading_session/frame_000001.jpg | [
{
"from": "human",
"value": "<image>\nDescribe what the person is doing."
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: idle; tool: n/a; attention: away]"
}
] |
reading_session_2 | reading_session/frame_000002.jpg | [
{
"from": "human",
"value": "<image>\nWhat activity and tools are visible here?"
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: idle; tool: n/a; attention: away]"
}
] |
reading_session_3 | reading_session/frame_000003.jpg | [
{
"from": "human",
"value": "<image>\nWhat is happening in this scene?"
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: idle; tool: n/a; attention: away]"
}
] |
reading_session_4 | reading_session/frame_000004.jpg | [
{
"from": "human",
"value": "<image>\nDescribe what the person is doing."
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: idle; tool: n/a; attention: away]"
}
] |
reading_session_5 | reading_session/frame_000005.jpg | [
{
"from": "human",
"value": "<image>\nWhat activity and tools are visible here?"
},
{
"from": "gpt",
"value": "C objects: bottle, pen; reading a page\n[verb: idle; tool: n/a; attention: away]"
}
] |
reading_session_6 | reading_session/frame_000006.jpg | [
{
"from": "human",
"value": "<image>\nWhat is happening in this scene?"
},
{
"from": "gpt",
"value": "C objects: bottle, pen; activity: leave_desk; reading a page\n[verb: leave; tool: n/a; attention: away]"
}
] |
reading_session_7 | reading_session/frame_000007.jpg | [
{
"from": "human",
"value": "<image>\nDescribe what the person is doing."
},
{
"from": "gpt",
"value": "C objects: bottle, pen; activity: leave_desk; reading a page\n[verb: leave; tool: n/a; attention: away]"
}
] |
corbanyax/pca-desk-starter
Personal desk/screen activity dataset captured and labelled with PCA — Personal Cognitive Assistant. Everything was captured locally and labelled on-device; only this exported, redacted dataset was uploaded.
| Stat | Value |
|---|---|
| Episodes | 2 |
| Frames | 16 |
| FPS (nominal) | 2.0 |
| Format | LeRobot v3.0 (+ LLaVA / Ego4D / COCO exports) |
What's inside
LeRobot v3 layout — meta/ (info, tasks, episodes), data/chunk-000/*.parquet
(per-frame timesteps), media/ (frames). Each timestep carries a frozen-taxonomy
tag: verb, nouns, tool, intent, modality, attention, narration.
Stream it
from huggingface_hub import snapshot_download
local = snapshot_download(repo_id="corbanyax/pca-desk-starter", repo_type="dataset")
# parquet timesteps — stream without loading everything
import pyarrow.dataset as ds
table = ds.dataset(f"{local}/data", format="parquet")
print(table.to_table().to_pandas().head())
Or with the 🤗 datasets streaming API:
from datasets import load_dataset
d = load_dataset("corbanyax/pca-desk-starter", split="train", streaming=True)
for row in d:
print(row["narration"], row["verb"], row["nouns"])
break
Reproduce / extend
pip install "pca[vision,dataset] @ git+https://github.com/atharvavdeo/pca"
pca-convert ./my_videos ./out --backend auto # auto-picks model by your RAM
pca-convert ./my_videos ./out --push corbanyax/pca-desk-starter # convert + upload
Privacy
Text was passed through PCA's PII redactor before upload (API keys, JWTs, emails, cards, phone/IP/IBAN, etc. masked). Review frames before publishing a public repo.
Card generated by pca.dataset.hub.
- Downloads last month
- 55