Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -1,17 +1,145 @@
|
|
| 1 |
---
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
language:
|
| 6 |
- en
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
tags:
|
| 8 |
- data-preprocessing
|
| 9 |
- automl
|
| 10 |
- quality-issues
|
| 11 |
- benchmarks
|
| 12 |
-
|
| 13 |
-
-
|
| 14 |
-
- 10K<n<100K
|
| 15 |
---
|
| 16 |
|
| 17 |
# Data Preprocessing AutoML Benchmarks
|
|
@@ -24,9 +152,8 @@ This repository contains text classification datasets with known data quality is
|
|
| 24 |
- **ag_news**: News categorization with topic overlap
|
| 25 |
- **twenty_newsgroups**: Newsgroup posts with cross-posting
|
| 26 |
|
| 27 |
-
### Class Imbalance Issues
|
| 28 |
- **yelp_polarity**: Sentiment analysis with rating bias
|
| 29 |
-
- **sms_spam**: Spam detection with severe imbalance
|
| 30 |
|
| 31 |
### Label Noise Issues
|
| 32 |
- **imdb**: Movie reviews with subjective labels
|
|
@@ -34,10 +161,6 @@ This repository contains text classification datasets with known data quality is
|
|
| 34 |
|
| 35 |
### Outlier Issues
|
| 36 |
- **emotion**: Twitter emotion with length outliers
|
| 37 |
-
- **financial_phrasebank**: Financial sentiment with domain outliers
|
| 38 |
-
|
| 39 |
-
### Clean Baselines
|
| 40 |
-
- **trec**: Question classification with clean labels
|
| 41 |
|
| 42 |
## Dataset Structure
|
| 43 |
|
|
@@ -62,11 +185,11 @@ dataset = load_dataset("MothMalone/data-preprocessing-automl-benchmarks", "ag_ne
|
|
| 62 |
|
| 63 |
# Access splits
|
| 64 |
train_data = dataset["train"]
|
| 65 |
-
val_data = dataset["validation"]
|
| 66 |
test_data = dataset["test"]
|
| 67 |
```
|
| 68 |
|
| 69 |
-
##
|
| 70 |
|
| 71 |
ag_news:
|
| 72 |
class_names:
|
|
|
|
| 1 |
---
|
| 2 |
+
configs:
|
| 3 |
+
- config_name: ag_news
|
| 4 |
+
data_files:
|
| 5 |
+
- path: ag_news/train.csv
|
| 6 |
+
split: train
|
| 7 |
+
- path: ag_news/validation.csv
|
| 8 |
+
split: validation
|
| 9 |
+
- path: ag_news/test.csv
|
| 10 |
+
split: test
|
| 11 |
+
- config_name: amazon_polarity
|
| 12 |
+
data_files:
|
| 13 |
+
- path: amazon_polarity/train.csv
|
| 14 |
+
split: train
|
| 15 |
+
- path: amazon_polarity/validation.csv
|
| 16 |
+
split: validation
|
| 17 |
+
- path: amazon_polarity/test.csv
|
| 18 |
+
split: test
|
| 19 |
+
- config_name: emotion
|
| 20 |
+
data_files:
|
| 21 |
+
- path: emotion/train.csv
|
| 22 |
+
split: train
|
| 23 |
+
- path: emotion/validation.csv
|
| 24 |
+
split: validation
|
| 25 |
+
- path: emotion/test.csv
|
| 26 |
+
split: test
|
| 27 |
+
- config_name: imdb
|
| 28 |
+
data_files:
|
| 29 |
+
- path: imdb/train.csv
|
| 30 |
+
split: train
|
| 31 |
+
- path: imdb/validation.csv
|
| 32 |
+
split: validation
|
| 33 |
+
- path: imdb/test.csv
|
| 34 |
+
split: test
|
| 35 |
+
- config_name: twenty_newsgroups
|
| 36 |
+
data_files:
|
| 37 |
+
- path: twenty_newsgroups/train.csv
|
| 38 |
+
split: train
|
| 39 |
+
- path: twenty_newsgroups/validation.csv
|
| 40 |
+
split: validation
|
| 41 |
+
- path: twenty_newsgroups/test.csv
|
| 42 |
+
split: test
|
| 43 |
+
- config_name: yelp_polarity
|
| 44 |
+
data_files:
|
| 45 |
+
- path: yelp_polarity/train.csv
|
| 46 |
+
split: train
|
| 47 |
+
- path: yelp_polarity/validation.csv
|
| 48 |
+
split: validation
|
| 49 |
+
- path: yelp_polarity/test.csv
|
| 50 |
+
split: test
|
| 51 |
+
dataset_info:
|
| 52 |
+
- config_name: ag_news
|
| 53 |
+
features:
|
| 54 |
+
- dtype: string
|
| 55 |
+
name: text
|
| 56 |
+
- dtype: int64
|
| 57 |
+
name: label
|
| 58 |
+
splits:
|
| 59 |
+
- name: train
|
| 60 |
+
num_examples: 90000
|
| 61 |
+
- name: validation
|
| 62 |
+
num_examples: 30000
|
| 63 |
+
- name: test
|
| 64 |
+
num_examples: 7600
|
| 65 |
+
- config_name: amazon_polarity
|
| 66 |
+
features:
|
| 67 |
+
- dtype: string
|
| 68 |
+
name: text
|
| 69 |
+
- dtype: int64
|
| 70 |
+
name: label
|
| 71 |
+
splits:
|
| 72 |
+
- name: train
|
| 73 |
+
num_examples: 2700000
|
| 74 |
+
- name: validation
|
| 75 |
+
num_examples: 900000
|
| 76 |
+
- name: test
|
| 77 |
+
num_examples: 400000
|
| 78 |
+
- config_name: emotion
|
| 79 |
+
features:
|
| 80 |
+
- dtype: string
|
| 81 |
+
name: text
|
| 82 |
+
- dtype: int64
|
| 83 |
+
name: label
|
| 84 |
+
splits:
|
| 85 |
+
- name: train
|
| 86 |
+
num_examples: 250085
|
| 87 |
+
- name: validation
|
| 88 |
+
num_examples: 83362
|
| 89 |
+
- name: test
|
| 90 |
+
num_examples: 41681
|
| 91 |
+
- config_name: imdb
|
| 92 |
+
features:
|
| 93 |
+
- dtype: string
|
| 94 |
+
name: text
|
| 95 |
+
- dtype: int64
|
| 96 |
+
name: label
|
| 97 |
+
splits:
|
| 98 |
+
- name: train
|
| 99 |
+
num_examples: 18750
|
| 100 |
+
- name: validation
|
| 101 |
+
num_examples: 6250
|
| 102 |
+
- name: test
|
| 103 |
+
num_examples: 25000
|
| 104 |
+
- config_name: twenty_newsgroups
|
| 105 |
+
features:
|
| 106 |
+
- dtype: string
|
| 107 |
+
name: text
|
| 108 |
+
- dtype: int64
|
| 109 |
+
name: label
|
| 110 |
+
splits:
|
| 111 |
+
- name: train
|
| 112 |
+
num_examples: 8485
|
| 113 |
+
- name: validation
|
| 114 |
+
num_examples: 2829
|
| 115 |
+
- name: test
|
| 116 |
+
num_examples: 7532
|
| 117 |
+
- config_name: yelp_polarity
|
| 118 |
+
features:
|
| 119 |
+
- dtype: string
|
| 120 |
+
name: text
|
| 121 |
+
- dtype: int64
|
| 122 |
+
name: label
|
| 123 |
+
splits:
|
| 124 |
+
- name: train
|
| 125 |
+
num_examples: 420000
|
| 126 |
+
- name: validation
|
| 127 |
+
num_examples: 140000
|
| 128 |
+
- name: test
|
| 129 |
+
num_examples: 38000
|
| 130 |
language:
|
| 131 |
- en
|
| 132 |
+
license: apache-2.0
|
| 133 |
+
size_categories:
|
| 134 |
+
- 1K<n<10K
|
| 135 |
+
- 10K<n<100K
|
| 136 |
tags:
|
| 137 |
- data-preprocessing
|
| 138 |
- automl
|
| 139 |
- quality-issues
|
| 140 |
- benchmarks
|
| 141 |
+
task_categories:
|
| 142 |
+
- text-classification
|
|
|
|
| 143 |
---
|
| 144 |
|
| 145 |
# Data Preprocessing AutoML Benchmarks
|
|
|
|
| 152 |
- **ag_news**: News categorization with topic overlap
|
| 153 |
- **twenty_newsgroups**: Newsgroup posts with cross-posting
|
| 154 |
|
| 155 |
+
### Class Imbalance Issues
|
| 156 |
- **yelp_polarity**: Sentiment analysis with rating bias
|
|
|
|
| 157 |
|
| 158 |
### Label Noise Issues
|
| 159 |
- **imdb**: Movie reviews with subjective labels
|
|
|
|
| 161 |
|
| 162 |
### Outlier Issues
|
| 163 |
- **emotion**: Twitter emotion with length outliers
|
|
|
|
|
|
|
|
|
|
|
|
|
| 164 |
|
| 165 |
## Dataset Structure
|
| 166 |
|
|
|
|
| 185 |
|
| 186 |
# Access splits
|
| 187 |
train_data = dataset["train"]
|
| 188 |
+
val_data = dataset["validation"]
|
| 189 |
test_data = dataset["test"]
|
| 190 |
```
|
| 191 |
|
| 192 |
+
## Dataset Details
|
| 193 |
|
| 194 |
ag_news:
|
| 195 |
class_names:
|