IOAI26: train=4000 (40% noise) | valid=500 (clean, stratified 25%) | test=1500 (labels hidden, stratified 75%)

README.md

@@ -0,0 +1,29 @@
+---
+dataset_info:
+  features:
+  - name: text
+    dtype: string
+  - name: label
+    dtype: int64
+  splits:
+  - name: train
+    num_bytes: 434038
+    num_examples: 4000
+  - name: validation
+    num_bytes: 54966
+    num_examples: 500
+  - name: test
+    num_bytes: 159729
+    num_examples: 1500
+  download_size: 384195
+  dataset_size: 648733
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+  - split: validation
+    path: data/validation-*
+  - split: test
+    path: data/test-*
+---

data/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c83354db94c9e32c9122ec6ea4eae4ad2dfc766739ac19f3351a0f3b955732b
+size 94216

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:874eb61e6b7314e87014460d5c3ac84c3ea62baca007b3bb2ad5cfe0a5f24cb2
+size 256183

data/validation-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c55fdc3f1ef34cfaf7b3a4a3fd7bd5aae99d4820fd13112031147c40f5daa32b
+size 33796