---
title: Ad Classification Exploration
emoji: 📰
colorFrom: yellow
colorTo: blue
sdk: gradio
sdk_version: 6.10.0
app_file: app.py
pinned: false
license: cc-by-sa-4.0
short_description: Explore yearly ad and non-ad distributions in Impresso
---

The app loads this aggregated JSON file from the repo by default:

`content-item-classification-base-multilingual_v1-0-0_aggregated_for_exploration.json`

You can override the source with the `DATA_SOURCE` environment variable.
Supported values are local paths, `http(s)` URLs, and `s3://` URLs.

Expected row shape:

```json
[
  {
    "provider_alias": "SFA",
    "provider_name": "Swiss Federal Archives",
    "newspaper_alias": "FedGazDe",
    "newspaper_title": "Bundesblatt",
    "year": 1851,
    "ad_count": 1,
    "non_ad_count": 18,
    "total_count": 19,
    "ad_share": 0.0526
  }
]
```

Optional S3 env vars when `DATA_SOURCE` uses `s3://`:
- `AWS_ENDPOINT_URL` or `S3_ENDPOINT_URL`
- `AWS_REGION` or `S3_REGION`
- `AWS_PROFILE` or `S3_PROFILE`