Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

(please review and merge) Zhenhui fix the data description of sampled CRITEO #633

Merged
merged 2 commits into from
Mar 12, 2019
Merged

(please review and merge) Zhenhui fix the data description of sampled CRITEO #633

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6bd1725
Fix the number and download link of used CRITEO samples (1M -> 100K) …
motefly Mar 12, 2019
b58d023
Fix the number and download link of used CRITEO samples (1M -> 100K) …
motefly Mar 12, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
60 changes: 30 additions & 30 deletions notebooks/00_quick_start/lightgbm_tinycriteo.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -116,7 +116,7 @@
"metadata": {},
"source": [
"## Data Preparation\n",
"Here we use CSV format as the example data input. Our example data is a sample (about 1 million samples) from Criteo dataset [2]. The Criteo dataset is a well-known industry benchmarking dataset for developing CTR prediction models, and it's frequently adopted as evaluation dataset by research papers. The original dataset is too large for a lightweight demo, so we sample a small portion from it as a demo dataset. <br>\n",
"Here we use CSV format as the example data input. Our example data is a sample (about 100 thousand samples) from Criteo dataset [2]. The Criteo dataset is a well-known industry benchmarking dataset for developing CTR prediction models, and it's frequently adopted as evaluation dataset by research papers. The original dataset is too large for a lightweight demo, so we sample a small portion from it as a demo dataset. <br>\n",
"Specifically, there are 39 columns of features in Criteo, where 13 columns are numerical features (I1-I13) and the other 26 columns are categorical features (C1-C26)."
]
},
Expand All @@ -137,7 +137,7 @@
"name": "stdout",
"output_type": "stream",
"text": [
"Extracting component files from /tmp/tmp_it899gq/dac_sample.tar.gz.\n"
"Extracting component files from /tmp/tmpz0rodvbn/dac_sample.tar.gz.\n"
]
},
{
Expand Down Expand Up @@ -552,33 +552,33 @@
"name": "stderr",
"output_type": "stream",
"text": [
"2019-03-08 04:32:23,741 [INFO] Filtering and fillna features\n",
"100%|██████████████████████████████████████████████████████████████| 26/26 [00:01<00:00, 14.19it/s]\n",
"100%|█████████████████████████████████████████████████████████████| 13/13 [00:00<00:00, 664.07it/s]\n",
"2019-03-08 04:32:25,690 [INFO] Ordinal encoding cate features\n",
"2019-03-08 04:32:27,060 [INFO] Target encoding cate features\n",
"100%|██████████████████████████████████████████████████████████████| 26/26 [00:03<00:00, 6.66it/s]\n",
"2019-03-08 04:32:30,974 [INFO] Start manual binary encoding\n",
"100%|██████████████████████████████████████████████████████████████| 65/65 [00:03<00:00, 16.17it/s]\n",
"100%|██████████████████████████████████████████████████████████████| 26/26 [00:02<00:00, 8.43it/s]\n",
"2019-03-08 04:32:37,119 [INFO] Filtering and fillna features\n",
"100%|█████████████████████████████████████████████████████████████| 26/26 [00:00<00:00, 166.30it/s]\n",
"100%|████████████████████████████████████████████████████████████| 13/13 [00:00<00:00, 1813.00it/s]\n",
"2019-03-08 04:32:37,286 [INFO] Ordinal encoding cate features\n",
"2019-03-08 04:32:37,451 [INFO] Target encoding cate features\n",
"100%|██████████████████████████████████████████████████████████████| 26/26 [00:00<00:00, 52.52it/s]\n",
"2019-03-08 04:32:37,948 [INFO] Start manual binary encoding\n",
"100%|██████████████████████████████████████████████████████████████| 65/65 [00:02<00:00, 26.73it/s]\n",
"100%|██████████████████████████████████████████████████████████████| 26/26 [00:01<00:00, 22.91it/s]\n",
"2019-03-08 04:32:41,597 [INFO] Filtering and fillna features\n",
"100%|█████████████████████████████████████████████████████████████| 26/26 [00:00<00:00, 168.82it/s]\n",
"100%|████████████████████████████████████████████████████████████| 13/13 [00:00<00:00, 2312.77it/s]\n",
"2019-03-08 04:32:41,761 [INFO] Ordinal encoding cate features\n",
"2019-03-08 04:32:41,922 [INFO] Target encoding cate features\n",
"100%|██████████████████████████████████████████████████████████████| 26/26 [00:00<00:00, 52.25it/s]\n",
"2019-03-08 04:32:42,422 [INFO] Start manual binary encoding\n",
"100%|██████████████████████████████████████████████████████████████| 65/65 [00:02<00:00, 26.44it/s]\n",
"100%|██████████████████████████████████████████████████████████████| 26/26 [00:01<00:00, 23.01it/s]"
"2019-03-12 07:55:56,415 [INFO] Filtering and fillna features\n",
"100%|██████████| 26/26 [00:02<00:00, 12.67it/s]\n",
"100%|██████████| 13/13 [00:00<00:00, 604.84it/s]\n",
"2019-03-12 07:55:58,494 [INFO] Ordinal encoding cate features\n",
"2019-03-12 07:55:59,943 [INFO] Target encoding cate features\n",
"100%|██████████| 26/26 [00:03<00:00, 6.48it/s]\n",
"2019-03-12 07:56:03,878 [INFO] Start manual binary encoding\n",
"100%|██████████| 65/65 [00:03<00:00, 16.50it/s]\n",
"100%|██████████| 26/26 [00:02<00:00, 7.86it/s]\n",
"2019-03-12 07:56:10,790 [INFO] Filtering and fillna features\n",
"100%|██████████| 26/26 [00:00<00:00, 167.89it/s]\n",
"100%|██████████| 13/13 [00:00<00:00, 1874.97it/s]\n",
"2019-03-12 07:56:10,956 [INFO] Ordinal encoding cate features\n",
"2019-03-12 07:56:11,120 [INFO] Target encoding cate features\n",
"100%|██████████| 26/26 [00:00<00:00, 52.55it/s]\n",
"2019-03-12 07:56:11,618 [INFO] Start manual binary encoding\n",
"100%|██████████| 65/65 [00:03<00:00, 21.43it/s]\n",
"100%|██████████| 26/26 [00:01<00:00, 18.60it/s]\n",
"2019-03-12 07:56:16,094 [INFO] Filtering and fillna features\n",
"100%|██████████| 26/26 [00:00<00:00, 151.00it/s]\n",
"100%|██████████| 13/13 [00:00<00:00, 2127.93it/s]\n",
"2019-03-12 07:56:16,288 [INFO] Ordinal encoding cate features\n",
"2019-03-12 07:56:16,453 [INFO] Target encoding cate features\n",
"100%|██████████| 26/26 [00:00<00:00, 52.20it/s]\n",
"2019-03-12 07:56:16,953 [INFO] Start manual binary encoding\n",
"100%|██████████| 65/65 [00:03<00:00, 21.40it/s]\n",
"100%|██████████| 26/26 [00:01<00:00, 18.42it/s]"
]
},
{
Expand Down Expand Up @@ -777,7 +777,7 @@
"source": [
"## Reference\n",
"\\[1\\] Guolin Ke, Qi Meng, Thomas Finley, Taifeng Wang, Wei Chen, Weidong Ma, Qiwei Ye, and Tie-Yan Liu. 2017. LightGBM: A highly efficient gradient boosting decision tree. In Advances in Neural Information Processing Systems. 3146–3154.<br>\n",
"\\[2\\] The Criteo datasets: http://labs.criteo.com/2014/02/kaggle-display-advertising-challenge-dataset/ .<br>\n",
"\\[2\\] The Criteo datasets: http://labs.criteo.com/wp-content/uploads/2015/04/dac_sample.tar.gz .<br>\n",
"\\[3\\] Anna Veronika Dorogush, Vasily Ershov, and Andrey Gulin. 2018. CatBoost: gradient boosting with categorical features support. arXiv preprint arXiv:1810.11363 (2018).<br>\n",
"\\[4\\] Scikit-learn. 2018. categorical_encoding. https://github.com/scikit-learn-contrib/categorical-encoding .<br>\n",
"\\[5\\] The parameters of LightGBM: https://github.com/Microsoft/LightGBM/blob/master/docs/Parameters.rst ."
Expand Down