Spaces:
Running
Running
Add CMD+RVL organization card
Browse files
README.md
CHANGED
|
@@ -1,10 +1,78 @@
|
|
| 1 |
---
|
| 2 |
title: README
|
| 3 |
-
emoji:
|
| 4 |
-
colorFrom:
|
| 5 |
-
colorTo:
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
title: README
|
| 3 |
+
emoji: π
|
| 4 |
+
colorFrom: indigo
|
| 5 |
+
colorTo: blue
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
+
# CMD+RVL β Structured finance data intelligence
|
| 11 |
+
|
| 12 |
+
We turn primary sources β SEC filings and the disclosures behind structured
|
| 13 |
+
finance deals β into machine-readable data with the provenance kept intact, so
|
| 14 |
+
every number can be traced back to where it came from.
|
| 15 |
+
|
| 16 |
+
The datasets on this page are **open samplers** of that work. They're built to be
|
| 17 |
+
downloaded, modeled, and pulled apart. The full, continuously-updated data that
|
| 18 |
+
backs them is available on the
|
| 19 |
+
[Snowflake](https://app.snowflake.com/marketplace/providers/GZT0ZYQ5618/CMD%2BRVL%20%28Command%20Reveal%29)
|
| 20 |
+
and [Databricks](https://marketplace.databricks.com/?provider=CMD%2BRVL&sortBy=date)
|
| 21 |
+
marketplaces β these Hugging Face releases are the taste, not the whole table.
|
| 22 |
+
|
| 23 |
+
## Explore the data live β [dealcharts.org](https://dealcharts.org)
|
| 24 |
+
|
| 25 |
+
DealCharts is our public window into structured finance: thousands of CMBS deals,
|
| 26 |
+
mapped funds, and BDCs, each backed by the SEC filings it came from. **Ask Cairn**
|
| 27 |
+
anything about a deal, fund, or entity and get an answer with its sources. Start
|
| 28 |
+
there β it's the fastest way to see what this data can do.
|
| 29 |
+
|
| 30 |
+
## On Hugging Face
|
| 31 |
+
|
| 32 |
+
**Available now β CMBS special-servicing early warning.** A leakage-safe
|
| 33 |
+
population framed around "which CMBS assets are at elevated risk of transferring
|
| 34 |
+
to special servicing within the next 12 months," using only point-in-time
|
| 35 |
+
information from SEC filings, in three shapes:
|
| 36 |
+
|
| 37 |
+
- [`cmbs-special-servicing-transfer`](https://huggingface.co/datasets/commandreveal/cmbs-special-servicing-transfer)
|
| 38 |
+
β flat supervised table for tabular classification, AutoML, XGBoost, and
|
| 39 |
+
tabular foundation models like [TabPFN](https://github.com/PriorLabs/TabPFN)
|
| 40 |
+
and [TabFM](https://github.com/google-research/tabfm).
|
| 41 |
+
- [`cmbs-special-servicing-sequences`](https://huggingface.co/datasets/commandreveal/cmbs-special-servicing-sequences)
|
| 42 |
+
β the same population as ordered per-asset sequences, for sequence and
|
| 43 |
+
time-series models.
|
| 44 |
+
- [`cmbs-special-servicing-timeline`](https://huggingface.co/datasets/commandreveal/cmbs-special-servicing-timeline)
|
| 45 |
+
β a text companion with a natural-language timeline per observation, for LLM
|
| 46 |
+
prompting, NLP, and text classification.
|
| 47 |
+
|
| 48 |
+
**What's next.** This is one use case in one asset class. More CMBS use cases,
|
| 49 |
+
Auto ABS, CLO, and other structured finance datasets are on the way.
|
| 50 |
+
|
| 51 |
+
## How to use these
|
| 52 |
+
|
| 53 |
+
Treat them as starting points, not answers. CMBS is hard, and these tables are
|
| 54 |
+
unlikely to be predictive on their own β even cleaned up. They're more
|
| 55 |
+
interesting as the *shape* of a problem: a defined population, an honest temporal
|
| 56 |
+
split, and traceable provenance to build on. Stitch your own signals onto the
|
| 57 |
+
spine β rates, spreads, macro, property-level data, your own features β and see
|
| 58 |
+
what actually moves the label.
|
| 59 |
+
|
| 60 |
+
The point of the provenance is that you never have to take a number on faith:
|
| 61 |
+
every observation carries the SEC filing and reporting period it came from, so
|
| 62 |
+
you can trace and check any value yourself. We're not claiming the labels are
|
| 63 |
+
clean or the feature set complete β we're claiming they're auditable. That's the
|
| 64 |
+
same discipline we bring to client work: finished results with proof that survive
|
| 65 |
+
without us.
|
| 66 |
+
|
| 67 |
+
## Terms
|
| 68 |
+
|
| 69 |
+
License: CC-BY-NC-4.0 β open for non-commercial use. For commercial use, get the
|
| 70 |
+
full data on
|
| 71 |
+
[Snowflake](https://app.snowflake.com/marketplace/providers/GZT0ZYQ5618/CMD%2BRVL%20%28Command%20Reveal%29)
|
| 72 |
+
or [Databricks](https://marketplace.databricks.com/?provider=CMD%2BRVL&sortBy=date),
|
| 73 |
+
or reach us at **cairn@cmdrvl.com**.
|
| 74 |
+
|
| 75 |
+
[dealcharts.org](https://dealcharts.org) Β·
|
| 76 |
+
[Snowflake Marketplace](https://app.snowflake.com/marketplace/providers/GZT0ZYQ5618/CMD%2BRVL%20%28Command%20Reveal%29) Β·
|
| 77 |
+
[Databricks Marketplace](https://marketplace.databricks.com/?provider=CMD%2BRVL&sortBy=date) Β·
|
| 78 |
+
[cmdrvl.com](https://cmdrvl.com) Β· cairn@cmdrvl.com
|