From 63471dec2ae4708b096dc9b9196d364dbb82817f Mon Sep 17 00:00:00 2001
From: jeffbrennan <jeffbrennan10@gmail.com>
Date: Tue, 14 Nov 2023 18:37:22 -0500
Subject: [PATCH] revert to current readme

---
 README.md | 22 ----------------------
 1 file changed, 22 deletions(-)

diff --git a/README.md b/README.md
index 7296846..e2c268e 100644
--- a/README.md
+++ b/README.md
@@ -114,25 +114,3 @@ Here's how to create a Parquet file with fake data:
 df = farsante.pandas_df([person.full_name, person.email, address.city, address.state, datetime.datetime], 3)
 df.to_parquet('./tmp/fake_data.parquet', index=False)
 ```
-
-## h2o data generation
-
-[h2o](https://github.com/h2oai/db-benchmark) is a popular dataset to benchmark data transformation performance. The main.rs file in [h2o-data-rust](h2o-data-rust/) can be used to generate h2o data.
-
-### arguments
-
-flag | description
-:--- | :---
---n | number of rows
---k | number of groups
---nas | percentage of missing values
---seed | the seed to use
-
-Here is a command to create a 1 million row dataset with 100 groups, 10% missing values, and a seed of 42.
-
-```bash
-cd h2o-data-rust
-cargo run -- --n 1000000 --k 100 --nas 10 --seed 42
-```
-
-A .csv file named G_1e6_1e2_10.csv will be created in the h2o-data-rust directory.