Added example of loading data in Datasets

README.md

@@ -38,7 +38,7 @@ BeanCounter is a dataset of business-oriented text. The data is sourced from the
 - `sample`: 1.1B tokens randomly sampled from `default` stratified by year
 
 ## How can I use this?
-
+### License
 The dataset is provided under the [ODC-By](https://opendatacommons.org/licenses/by/1-0/) license. Cite our work as:
 
 ```
@@ -51,3 +51,20 @@ The dataset is provided under the [ODC-By](https://opendatacommons.org/licenses/
     year={2024}
 }
 ```
+### In Hugging Face Datasets
+To load the random sample config in Datasets, one can run:
+```python
+from datasets import load_dataset
+
+beancounter = load_dataset(
+    "blevy41/BeanCounter",
+    name="sample", # Load random sample, clean, or final
+)
+
+# Print out split info
+print(beancounter, '\n')
+
+# Inspect an observation
+print(f"COLUMNS IN DATA: {','.join(beancounter['train'][1000].keys())}\n")
+print(f"EXCERPT: \n\n{beancounter['train'][1000]['text'][:1000]}")
+```