Upload README.md with huggingface_hub

README.md

@@ -1,107 +1,3 @@
----
-configs:
-- config_name: default
-  data_files:
-  - split: train_2
-    path: data/train_2-*
-  - split: test_2
-    path: data/test_2-*
-  - split: train_4
-    path: data/train_4-*
-  - split: test_4
-    path: data/test_4-*
-  - split: train_6
-    path: data/train_6-*
-  - split: test_6
-    path: data/test_6-*
-  - split: train_8
-    path: data/train_8-*
-  - split: test_8
-    path: data/test_8-*
-  - split: train_10
-    path: data/train_10-*
-  - split: test_10
-    path: data/test_10-*
-  - split: train_20
-    path: data/train_20-*
-  - split: test_20
-    path: data/test_20-*
-  - split: train_30
-    path: data/train_30-*
-  - split: test_30
-    path: data/test_30-*
-  - split: train_50
-    path: data/train_50-*
-  - split: test_50
-    path: data/test_50-*
-  - split: train_100
-    path: data/train_100-*
-  - split: test_100
-    path: data/test_100-*
-dataset_info:
-  features:
-  - name: input
-    sequence: string
-  - name: output
-    dtype: string
-  splits:
-  - name: train_2
-    num_bytes: 443070
-    num_examples: 8000
-  - name: test_2
-    num_bytes: 110938
-    num_examples: 2000
-  - name: train_4
-    num_bytes: 677672
-    num_examples: 8000
-  - name: test_4
-    num_bytes: 169725
-    num_examples: 2000
-  - name: train_6
-    num_bytes: 929545
-    num_examples: 8000
-  - name: test_6
-    num_bytes: 232342
-    num_examples: 2000
-  - name: train_8
-    num_bytes: 1177415
-    num_examples: 8000
-  - name: test_8
-    num_bytes: 293720
-    num_examples: 2000
-  - name: train_10
-    num_bytes: 1427596
-    num_examples: 8000
-  - name: test_10
-    num_bytes: 356409
-    num_examples: 2000
-  - name: train_20
-    num_bytes: 2810319
-    num_examples: 8000
-  - name: test_20
-    num_bytes: 701379
-    num_examples: 2000
-  - name: train_30
-    num_bytes: 4194268
-    num_examples: 8000
-  - name: test_30
-    num_bytes: 1048790
-    num_examples: 2000
-  - name: train_50
-    num_bytes: 6958734
-    num_examples: 8000
-  - name: test_50
-    num_bytes: 1738125
-    num_examples: 2000
-  - name: train_100
-    num_bytes: 13862843
-    num_examples: 8000
-  - name: test_100
-    num_bytes: 3462772
-    num_examples: 2000
-  download_size: 11904704
-  dataset_size: 40595662
----
 
 # Arithmetic Puzzles Dataset
 
@@ -113,10 +9,10 @@ Outputs are filtered to be between [-100, 100], and self-reference/looped depend
 
 Splits are named like:
 
-- `train_N` 10k total examples of puzzles with up to N variables
-- `test_N` 10k more examples with up to N variables
+- `train_N` 8k total examples of puzzles with N variables
+- `test_N` 2k more examples with N variables
 
-Note: presently, train/test leakage is not prevented. A training example may randomly be generated in the test set, and this is not filtered out.
+Train/test leakage is prevented: all training examples are filtered out of the test set.
 
 Conceptually the data looks like this:
 
@@ -148,11 +44,11 @@ In actuality it looks like this:
 from datasets import load_dataset
 
 # Load the entire dataset
-dataset = load_dataset("neurallambda/arithmetic_puzzles")
+dataset = load_dataset("neurallambda/arithmetic_dataset")
 
 # Load specific splits
-train_small = load_dataset("neurallambda/arithmetic_puzzles", split="train_10")
-test_small = load_dataset("neurallambda/arithmetic_puzzles", split="test_10")
+train_small = load_dataset("neurallambda/arithmetic_dataset", split="train_10")
+test_small = load_dataset("neurallambda/arithmetic_dataset", split="test_10")
 ```
 
 ### Preparing Inputs