Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code: DatasetGenerationError
Exception: UnicodeDecodeError
Message: 'utf-8' codec can't decode byte 0xe9 in position 3: invalid continuation byte
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1827, in _prepare_split_single
for key, table in generator:
^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 613, in wrapped
for item in generator(*args, **kwargs):
~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/text/text.py", line 98, in _generate_tables
batch = f.read(self.config.chunksize)
File "/usr/local/lib/python3.14/site-packages/datasets/utils/file_utils.py", line 839, in read_with_retries
out = read(*args, **kwargs)
File "<frozen codecs>", line 325, in decode
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xe9 in position 3: invalid continuation byte
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
~~~~~~~~~~~~~~~~~~~~~~~~~^
builder, max_dataset_size_bytes=max_dataset_size_bytes
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1694, in _prepare_split
for job_id, done, content in self._prepare_split_single(
~~~~~~~~~~~~~~~~~~~~~~~~~~^
gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
):
^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1879, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
text string |
|---|
--- |
pretty_name: HeadWater AI Data Ingestion Engine - Free Sample |
license: mit |
language: |
- en |
tags: |
- data-cleaning |
- data-ingestion |
- text-preprocessing |
- preprocessing |
- json-report |
- sample |
size_categories: |
- n<1K |
viewer: false |
--- |
# HeadWater AI Data Ingestion Engine: Free Sample |
This is a free, runnable sample of what the HeadWater AI Data Ingestion Engine does to messy input files. It has a set of deliberately messy sample files, the real output the full engine produced from them, the JSON report, and a small demo script you can run yourself. |
The full engine is a separate paid product and its code is not in this repo. |
## What the engine does |
You point it at a file or a folder. It: |
1. Finds every file in the folder, including subfolders. |
2. Cleans up text files: line endings, extra spaces and tabs, long runs of blank lines. |
3. Strips company suffixes like Inc., LLC and Corp. from names. |
4. Standardizes system parameter lines like `system parameter max retries = 5` into `SYSTEM_PARAMETER_MAX_RETRIES=5`. |
5. Flags text that looks like instructions aimed at an AI (for example "ignore previous instructions") and hidden or invisible characters. It flags them for you to look at and does not delete them. |
6. Writes the cleaned files plus a JSON report that lists every file, what happened to it, and why anything failed. One bad file does not stop the run. |
Your original files are never changed. |
## What's in this repo |
| Path | What it is | |
| --- | --- | |
| `sample_input/` | 13 messy files: Windows line endings, odd spacing, bad encoding, an empty file, a fake PDF, hidden characters, a prompt-injection line, duplicates in subfolders | |
| `engine_output/` | Real output from the full engine on `sample_input/`, including `reports/module_01_report.json` | |
| `demo_lite.py` | A small script you can run. It does the line-ending and spacing cleanup, flags suspicious text and hidden characters, and writes a report. It does not do the company-name or system-parameter steps | |
| `demo_output/` | What `demo_lite.py` produces | |
| `LICENSE` | MIT, for the sample files and the demo script only | |
## Run the demo |
You need Python 3.8 or newer. Nothing else to install. |
``` |
python3 demo_lite.py sample_input --output demo_output |
``` |
It prints a one-line summary and writes `demo_output/reports/demo_report.json`. You can also point it at your own folder of text files. |
## Before and after (real engine output) |
Input `01_meeting_notes_crlf.txt` (Windows line endings, tabs, extra spaces, blank lines): |
``` |
Weekly sync notes |
Attendees: Dana, Luis, Priya |
Action items: follow up with Northwind Traders Inc. and Blue Harbor LLC |
``` |
Output: |
``` |
Weekly sync notes |
Attendees: Dana, Luis, Priya |
Action items: follow up with Northwind Traders and Blue Harbor |
``` |
Input `03_config_dump.txt`: |
``` |
system parameter max retries = 5 |
SYSTEM-PARAM-timeout = 30s |
System Parameter log level=debug |
region = us-west |
``` |
Output (the `region` line is left alone because it is not a system parameter): |
``` |
SYSTEM_PARAMETER_MAX_RETRIES=5 |
SYSTEM_PARAM_TIMEOUT=30s |
SYSTEM_PARAMETER_LOG_LEVEL=debug |
region = us-west |
``` |
## The JSON report |
Every run writes a report. The top of the report from this sample: |
```json |
End of preview.
HeadWater AI Data Ingestion Engine
A standalone Python tool that gets incoming files ready for AI ingestion workflows. It runs your files through a defined workflow, prepares file and structure information for downstream use, and reports what it finds. You run it the same way every time new data comes in, so the work never starts over.
Built for engineers and data teams who have to get raw files ready for AI before anything else can happen. Version 1.0.
For HeadWater company and product information click here
Free Sample Download
HeadWater provides a downloadable free sample of the AI Data Ingestion Engine so prospective customers can examine the product package before considering the full commercial release.
Download: "Headwater AI Data Ingestion Engine Free Sample.zip"
How to Get Started
- Download the ZIP file from the repository's file list.
- Extract the ZIP into a new folder on your computer.
- Open the extracted folder and read the included documentation and license terms.
- Follow the package's instructions to run the sample, if executable components and usage instructions are included.
Important Scope Note
The free sample is intended for evaluation and demonstration. Its contents and capabilities may differ from the complete commercial product. Review the included files and documentation to determine the exact functionality available in this sample.
HeadWater develops and sells AI digital products designed to solve real industry problems.
- Downloads last month
- 90