Add README.md

README.md

@@ -1,96 +1,183 @@
 ---
+annotations_creators:
+- no-annotation
+language_creators:
+- found
+language:
+- en
+license: cc-by-4.0
+multilinguality:
+- monolingual
+size_categories:
+- 10K<n<100K
+source_datasets:
+- original
+task_categories:
+- tabular-classification
+- tabular-regression
+task_ids: []
+tags:
+- africa
+- humanitarian
+- hdx
+- electric-sheep-africa
+- eastern-africa
+- trade
+- tza
+pretty_name: "United Republic of Tanzania Daily FEWS NET Cross Border Trade Data"
 dataset_info:
-  features:
-  - name: reporting_country
-    dtype: string
-  - name: reporting_country_code
-    dtype: string
-  - name: border_point
-    dtype: string
-  - name: source
-    dtype: string
-  - name: source_country_code
-    dtype: string
-  - name: destination
-    dtype: string
-  - name: destination_country_code
-    dtype: string
-  - name: cpcv2
-    dtype: string
-  - name: product
-    dtype: string
-  - name: indicator_name
-    dtype: string
-  - name: start_date
-    dtype: timestamp[ns]
-  - name: period_date
-    dtype: timestamp[ns]
-  - name: value
-    dtype: float64
-  - name: flow_type
-    dtype: string
-  - name: trade_type
-    dtype: string
-  - name: collection_status
-    dtype: string
-  - name: source_organization
-    dtype: string
-  - name: source_document
-    dtype: string
-  - name: dataseries_name
-    dtype: string
-  - name: dataseries
-    dtype: int64
-  - name: unit
-    dtype: string
-  - name: unit_type
-    dtype: string
-  - name: unit_name
-    dtype: string
-  - name: status
-    dtype: string
-  - name: common_unit
-    dtype: string
-  - name: common_unit_quantity
-    dtype: float64
-  - name: reporting_country_geographic_group
-    dtype: string
-  - name: reporting_country_fewsnet_region
-    dtype: string
-  - name: source_geographic_group
-    dtype: string
-  - name: source_fewsnet_region
-    dtype: string
-  - name: destination_geographic_group
-    dtype: string
-  - name: destination_fewsnet_region
-    dtype: string
-  - name: value_one_month_ago
-    dtype: float64
-  - name: pct_change_from_one_month_ago
-    dtype: float64
-  - name: collection_schedule
-    dtype: string
-  - name: data_usage_policy
-    dtype: string
-  - name: esa_source
-    dtype: string
-  - name: esa_processed
-    dtype: string
   splits:
-  - name: train
-    num_bytes: 21975144
-    num_examples: 34144
-  - name: test
-    num_bytes: 5487220
-    num_examples: 8536
-  download_size: 2024411
-  dataset_size: 27462364
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
-  - split: test
-    path: data/test-*
+    - name: train
+      num_examples: 34144
+    - name: test
+      num_examples: 8536
 ---
+
+# United Republic of Tanzania Daily FEWS NET Cross Border Trade Data
+
+**Publisher:** FEWS NET · **Source:** [HDX](https://data.humdata.org/dataset/daily_cross_border_trade_for_united_republic_of_tanzania_6824) · **License:** `cc-by` · **Updated:** 2026-03-28
+
+---
+
+## Abstract
+
+United Republic of Tanzania Daily cross border trade data collected by FEWS NET since 2010.
+
+Each row in this dataset represents first-level administrative unit observations. Temporal coverage is indicated by the `start_date`, `period_date` column(s). Geographic scope: **TZA**.
+
+*Curated into ML-ready Parquet format by [Electric Sheep Africa](https://huggingface.co/electricsheepafrica).*
+
+---
+
+## Dataset Characteristics
+
+| | |
+|---|---|
+| **Domain** | Humanitarian and development data |
+| **Unit of observation** | First-level administrative unit observations |
+| **Rows (total)** | 42,680 |
+| **Columns** | 38 (5 numeric, 31 categorical, 2 datetime) |
+| **Train split** | 34,144 rows |
+| **Test split** | 8,536 rows |
+| **Geographic scope** | TZA |
+| **Publisher** | FEWS NET |
+| **HDX last updated** | 2026-03-28 |
+
+---
+
+## Variables
+
+**Geographic** — `reporting_country` (Tanzania, United Republic of, Kenya, Malawi), `reporting_country_code` (TZ, KE, MW), `source_country_code` (TZ, KE, BI), `destination_country_code` (KE, BI, TZ), `flow_type` and 8 others.
+
+**Temporal** — `start_date`, `period_date`, `value_one_month_ago` (range 0.2–83640500.0), `pct_change_from_one_month_ago` (range -100.0–40188788.8889).
+
+**Outcome / Measurement** — `value` (range 0.0–334562000.0).
+
+**Identifier / Metadata** — `source` (Tanzania, United Republic of, Kenya, Burundi), `indicator_name` (TradeFlowQuantity), `source_organization`, `source_document`, `dataseries_name` and 4 others.
+
+**Other** — `border_point` (Kibande, Taveta, Manyovu), `destination` (Kenya, Burundi, Tanzania, United Republic of), `cpcv2` (P23161AA, R01122AC, R01701AA), `product` (Rice (Milled), Maize Grain (White), Beans (mixed)), `collection_status` and 6 others.
+
+---
+
+## Quick Start
+
+```python
+from datasets import load_dataset
+
+ds    = load_dataset("electricsheepafrica/africa-daily-cross-border-trade-for-united-republic-of-tanzania-6824")
+train = ds["train"].to_pandas()
+test  = ds["test"].to_pandas()
+
+print(train.shape)
+train.head()
+```
+
+---
+
+## Schema
+
+| Column | Type | Null % | Range / Sample Values |
+|---|---|---|---|
+| `reporting_country` | object | 0.0% | Tanzania, United Republic of, Kenya, Malawi |
+| `reporting_country_code` | object | 0.0% | TZ, KE, MW |
+| `border_point` | object | 4.8% | Kibande, Taveta, Manyovu |
+| `source` | object | 0.0% | Tanzania, United Republic of, Kenya, Burundi |
+| `source_country_code` | object | 0.0% | TZ, KE, BI |
+| `destination` | object | 0.0% | Kenya, Burundi, Tanzania, United Republic of |
+| `destination_country_code` | object | 0.0% | KE, BI, TZ |
+| `cpcv2` | object | 0.0% | P23161AA, R01122AC, R01701AA |
+| `product` | object | 0.0% | Rice (Milled), Maize Grain (White), Beans (mixed) |
+| `indicator_name` | object | 0.0% | TradeFlowQuantity |
+| `start_date` | datetime64[ns] | 0.0% |  |
+| `period_date` | datetime64[ns] | 0.0% |  |
+| `value` | float64 | 0.0% | 0.0 – 334562000.0 (mean 64106.3632) |
+| `flow_type` | object | 0.0% |  |
+| `trade_type` | object | 0.0% |  |
+| `collection_status` | object | 0.0% |  |
+| `source_organization` | object | 0.0% |  |
+| `source_document` | object | 0.0% |  |
+| `dataseries_name` | object | 0.0% |  |
+| `dataseries` | int64 | 0.0% | 27989.0 – 7051434.0 (mean 6290835.0721) |
+| `unit` | object | 0.0% |  |
+| `unit_type` | object | 0.0% |  |
+| `unit_name` | object | 0.0% |  |
+| `status` | object | 0.0% |  |
+| `common_unit` | object | 0.0% |  |
+| `common_unit_quantity` | float64 | 0.0% | 0.0 – 334562000.0 (mean 279649.6762) |
+| `reporting_country_geographic_group` | object | 0.0% |  |
+| `reporting_country_fewsnet_region` | object | 0.0% |  |
+| `source_geographic_group` | object | 0.0% |  |
+| `source_fewsnet_region` | object | 0.0% |  |
+| `destination_geographic_group` | object | 0.0% |  |
+| `destination_fewsnet_region` | object | 0.0% |  |
+| `value_one_month_ago` | float64 | 76.7% | 0.2 – 83640500.0 (mean 63837.2234) |
+| `pct_change_from_one_month_ago` | float64 | 76.7% | -100.0 – 40188788.8889 (mean 5965.1906) |
+| `collection_schedule` | object | 0.0% |  |
+| `data_usage_policy` | object | 0.0% |  |
+| `esa_source` | object | 0.0% |  |
+| `esa_processed` | object | 0.0% |  |
+
+---
+
+## Numeric Summary
+
+| Column | Min | Max | Mean | Median |
+|---|---|---|---|---|
+| `value` | 0.0 | 334562000.0 | 64106.3632 | 0.0 |
+| `dataseries` | 27989.0 | 7051434.0 | 6290835.0721 | 6606785.0 |
+| `common_unit_quantity` | 0.0 | 334562000.0 | 279649.6762 | 0.0 |
+| `value_one_month_ago` | 0.2 | 83640500.0 | 63837.2234 | 274.25 |
+| `pct_change_from_one_month_ago` | -100.0 | 40188788.8889 | 5965.1906 | 215.6977 |
+
+---
+
+## Curation
+
+Raw data was downloaded from HDX via the CKAN API and converted to Parquet. Column names were lowercased and standardised to snake_case. Common missing-value markers (`N/A`, `null`, `none`, `-`, `unknown`, `no data`, `#N/A`) were unified to `NaN`. 10 column(s) with >80% missing values were removed: `id`, `value_one_year_ago`, `value_two_years_ago`, `value_three_years_ago`, `value_four_years_ago`, `value_five_years_ago`.... 2 column(s) were cast from string to numeric or datetime based on parse-success rate (>85% threshold). The dataset was split 80/20 into train and test partitions using a fixed random seed (42) and saved as Snappy-compressed Parquet.
+
+---
+
+## Limitations
+
+- Data originates from FEWS NET and has not been independently validated by ESA.
+- Automated cleaning cannot correct for misreported values, definitional inconsistencies, or sampling bias in the original collection.
+- The following columns have >20% missing values and should be treated with caution in modelling: `value_one_month_ago`, `pct_change_from_one_month_ago`.
+- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/daily_cross_border_trade_for_united_republic_of_tanzania_6824) for the publisher's own methodology notes and caveats.
+
+---
+
+## Citation
+
+```bibtex
+@dataset{hdx_africa_daily_cross_border_trade_for_united_republic_of_tanzania_6824,
+  title     = {United Republic of Tanzania Daily FEWS NET Cross Border Trade Data},
+  author    = {FEWS NET},
+  year      = {2026},
+  url       = {https://data.humdata.org/dataset/daily_cross_border_trade_for_united_republic_of_tanzania_6824},
+  note      = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
+}
+```
+
+---
+
+*[Electric Sheep Africa](https://huggingface.co/electricsheepafrica) — Africa's ML dataset infrastructure. Lagos, Nigeria.*