# Up to 4x faster than Series.str.contains / Series.eq - can handle Unicode!
```python
pip install a-pandas-ex-fast-string
```
```python
from a_pandas_ex_fast_string import pd_add_fast_string
import pandas as pd
pd_add_fast_string()
df2 = pd.read_csv(
"https://raw.githubusercontent.com/pandas-dev/pandas/main/doc/data/titanic.csv",
dtype="string",
)
# To check if it can handle unicode strings
df2.Name.iloc[0] += "ö"
df2.Name.iloc[10] += "ä"
df2.Name.iloc[20] += "ü"
# converts the whole dataframe
df900 = pd.Q_convert_to_fast_string(df2.copy())
dfone = df2.copy()
# converts one column
dfone.Cabin.ds_update_fast_string()
# Let's create some DataFrames of different sizes
df9000 = pd.Q_convert_to_fast_string(
pd.concat([df2.copy() for _ in range(10)], ignore_index=True)
)
df90000 = pd.Q_convert_to_fast_string(
pd.concat([df2.copy() for _ in range(100)], ignore_index=True)
)
df900000 = pd.Q_convert_to_fast_string(
pd.concat([df2.copy() for _ in range(1000)], ignore_index=True)
)
df9000000 = pd.Q_convert_to_fast_string(
pd.concat([df2.copy() for _ in range(10000)], ignore_index=True)
)
%timeit df900.loc[df900.Name.s_string_contains('y') | df900.Name.s_string_is('Montvila, Rev. Juozas')]
%timeit df900.loc[df900.Name.str.contains('y',regex=False) | (df900.Name == 'Montvila, Rev. Juozas')]
604 µs ± 9.09 µs per loop (mean ± std. dev. of 7 runs, 1,000 loops each)
997 µs ± 13.2 µs per loop (mean ± std. dev. of 7 runs, 1,000 loops each)
%timeit df9000.loc[df9000.Name.s_string_contains('y') | df9000.Name.s_string_is('Montvila, Rev. Juozas')]
%timeit df9000.loc[df9000.Name.str.contains('y',regex=False) | (df9000.Name == 'Montvila, Rev. Juozas')]
1.15 ms ± 15.2 µs per loop (mean ± std. dev. of 7 runs, 1,000 loops each)
2.77 ms ± 11.2 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)
%timeit df90000.loc[df90000.Name.s_string_contains('y') | df90000.Name.s_string_is('Montvila, Rev. Juozas')]
%timeit df90000.loc[df90000.Name.str.contains('y',regex=False) | (df90000.Name == 'Montvila, Rev. Juozas')]
6.45 ms ± 77.4 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)
20.7 ms ± 166 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)
%timeit df900000.loc[df900000.Name.s_string_contains('y') | df900000.Name.s_string_is('Montvila, Rev. Juozas')]
%timeit df900000.loc[df900000.Name.str.contains('y',regex=False) | (df900000.Name == 'Montvila, Rev. Juozas')]
60.5 ms ± 853 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)
206 ms ± 840 µs per loop (mean ± std. dev. of 7 runs, 1 loop each)
%timeit df9000000.loc[df9000000.Name.s_string_contains('y') | df9000000.Name.s_string_is('Montvila, Rev. Juozas')]
%timeit df9000000.loc[df9000000.Name.str.contains('y',regex=False) | (df9000000.Name == 'Montvila, Rev. Juozas')]
596 ms ± 11.8 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
2.06 s ± 2.5 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
# Good news: it can handle unicode characters!
df9000.loc[df9000.Name.s_string_contains('ö')].Name
Out[14]:
0 Braund, Mr. Owen Harrisö
891 Braund, Mr. Owen Harrisö
1782 Braund, Mr. Owen Harrisö
2673 Braund, Mr. Owen Harrisö
3564 Braund, Mr. Owen Harrisö
4455 Braund, Mr. Owen Harrisö
5346 Braund, Mr. Owen Harrisö
6237 Braund, Mr. Owen Harrisö
7128 Braund, Mr. Owen Harrisö
8019 Braund, Mr. Owen Harrisö
Name: Name, dtype: string
# Bad news: every time you modify a Series, you have to update it:
df9000.loc[df9000.Name.s_string_contains('ö')].Name
0 Braund, Mr. Owen Harrisö
891 Braund, Mr. Owen Harrisö
1782 Braund, Mr. Owen Harrisö
2673 Braund, Mr. Owen Harrisö
3564 Braund, Mr. Owen Harrisö
df9000.loc[df9000.Name.s_string_contains('ö'), "Name"] = df9000.loc[df9000.Name.s_string_contains('ö'), "Name"] + 'Ä' # updating
df9000.Name
0 Braund, Mr. Owen HarrisöÄ
1 Cumings, Mrs. John Bradley (Florence Briggs Th...
2 Heikkinen, Miss. Laina
df9000.loc[df9000.Name.s_string_contains('ö'), "Name"] # Exception because ds_update_fast_string was not called
Traceback (most recent call last):
File "C:\Users\Gamer\anaconda3\envs\dfdir\lib\site-packages\IPython\core\interactiveshell.py", line 3398, in run_code
exec(code_obj, self.user_global_ns, self.user_ns)
File "<ipython-input-7-2b0dfaf8b41c>", line 1, in <cell line: 1>
df9000.loc[df9000.Name.s_string_contains('ö'), "Name"]
File "C:/Users/Gamer/anaconda3/envs/dfdir/a_pandas_string_search.py", line 133, in search_contains
wordtosearchbin, columntosearch = _get_col_word(
File "C:/Users/Gamer/anaconda3/envs/dfdir/a_pandas_string_search.py", line 103, in _get_col_word
return wordtosearchbin, series._stringser.__array__()
AttributeError: 'NoneType' object has no attribute '__array__'
df9000.Name.ds_update_fast_string() # Necessary after changing a Series
# you can also update the whole DataFrame: df9000 = df9000.ds_update_fast_string()
# Be careful: df9000.Name.ds_update_fast_string() returns None (inplace)
# df9000.ds_update_fast_string() returns a DataFrame
df9000.loc[df9000.Name.s_string_contains('ö'), "Name"] # Now it is working!
0 Braund, Mr. Owen HarrisöÄ
891 Braund, Mr. Owen HarrisöÄ
1782 Braund, Mr. Owen HarrisöÄ
2673 Braund, Mr. Owen HarrisöÄ
```
Raw data
{
"_id": null,
"home_page": "https://github.com/hansalemaos/a_pandas_ex_fast_string",
"name": "a-pandas-ex-fast-string",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "pandas,DataFrame,string,Series,fast,unicode",
"author": "Johannes Fischer",
"author_email": "<aulasparticularesdealemaosp@gmail.com>",
"download_url": "https://files.pythonhosted.org/packages/05/74/0e7a4707e95520e37bd4f1e3ba8eceadcdd2a7ce78ceb25d633641cea9ff/a_pandas_ex_fast_string-0.11.tar.gz",
"platform": null,
"description": "\n# Up to 4x faster than Series.str.contains / Series.eq - can handle Unicode!\n\n\n\n```python\n\npip install a-pandas-ex-fast-string\n\n```\n\n\n\n```python\n\nfrom a_pandas_ex_fast_string import pd_add_fast_string\n\nimport pandas as pd\n\n\n\npd_add_fast_string()\n\n\n\ndf2 = pd.read_csv(\n\n \"https://raw.githubusercontent.com/pandas-dev/pandas/main/doc/data/titanic.csv\",\n\n dtype=\"string\",\n\n)\n\n\n\n# To check if it can handle unicode strings\n\ndf2.Name.iloc[0] += \"\u00f6\"\n\ndf2.Name.iloc[10] += \"\u00e4\"\n\ndf2.Name.iloc[20] += \"\u00fc\"\n\n\n\n# converts the whole dataframe\n\ndf900 = pd.Q_convert_to_fast_string(df2.copy())\n\n\n\n\n\ndfone = df2.copy()\n\n# converts one column\n\ndfone.Cabin.ds_update_fast_string()\n\n\n\n# Let's create some DataFrames of different sizes\n\ndf9000 = pd.Q_convert_to_fast_string(\n\n pd.concat([df2.copy() for _ in range(10)], ignore_index=True)\n\n)\n\ndf90000 = pd.Q_convert_to_fast_string(\n\n pd.concat([df2.copy() for _ in range(100)], ignore_index=True)\n\n)\n\ndf900000 = pd.Q_convert_to_fast_string(\n\n pd.concat([df2.copy() for _ in range(1000)], ignore_index=True)\n\n)\n\ndf9000000 = pd.Q_convert_to_fast_string(\n\n pd.concat([df2.copy() for _ in range(10000)], ignore_index=True)\n\n)\n\n\n\n\n\n\n\n%timeit df900.loc[df900.Name.s_string_contains('y') | df900.Name.s_string_is('Montvila, Rev. Juozas')]\n\n%timeit df900.loc[df900.Name.str.contains('y',regex=False) | (df900.Name == 'Montvila, Rev. Juozas')]\n\n604 \u00b5s \u00b1 9.09 \u00b5s per loop (mean \u00b1 std. dev. of 7 runs, 1,000 loops each)\n\n997 \u00b5s \u00b1 13.2 \u00b5s per loop (mean \u00b1 std. dev. of 7 runs, 1,000 loops each)\n\n\n\n\n\n%timeit df9000.loc[df9000.Name.s_string_contains('y') | df9000.Name.s_string_is('Montvila, Rev. Juozas')]\n\n%timeit df9000.loc[df9000.Name.str.contains('y',regex=False) | (df9000.Name == 'Montvila, Rev. Juozas')]\n\n1.15 ms \u00b1 15.2 \u00b5s per loop (mean \u00b1 std. dev. of 7 runs, 1,000 loops each)\n\n2.77 ms \u00b1 11.2 \u00b5s per loop (mean \u00b1 std. dev. of 7 runs, 100 loops each)\n\n\n\n\n\n%timeit df90000.loc[df90000.Name.s_string_contains('y') | df90000.Name.s_string_is('Montvila, Rev. Juozas')]\n\n%timeit df90000.loc[df90000.Name.str.contains('y',regex=False) | (df90000.Name == 'Montvila, Rev. Juozas')]\n\n6.45 ms \u00b1 77.4 \u00b5s per loop (mean \u00b1 std. dev. of 7 runs, 100 loops each)\n\n20.7 ms \u00b1 166 \u00b5s per loop (mean \u00b1 std. dev. of 7 runs, 10 loops each)\n\n\n\n\n\n%timeit df900000.loc[df900000.Name.s_string_contains('y') | df900000.Name.s_string_is('Montvila, Rev. Juozas')]\n\n%timeit df900000.loc[df900000.Name.str.contains('y',regex=False) | (df900000.Name == 'Montvila, Rev. Juozas')]\n\n60.5 ms \u00b1 853 \u00b5s per loop (mean \u00b1 std. dev. of 7 runs, 10 loops each)\n\n206 ms \u00b1 840 \u00b5s per loop (mean \u00b1 std. dev. of 7 runs, 1 loop each)\n\n\n\n\n\n%timeit df9000000.loc[df9000000.Name.s_string_contains('y') | df9000000.Name.s_string_is('Montvila, Rev. Juozas')]\n\n%timeit df9000000.loc[df9000000.Name.str.contains('y',regex=False) | (df9000000.Name == 'Montvila, Rev. Juozas')]\n\n596 ms \u00b1 11.8 ms per loop (mean \u00b1 std. dev. of 7 runs, 1 loop each)\n\n2.06 s \u00b1 2.5 ms per loop (mean \u00b1 std. dev. of 7 runs, 1 loop each)\n\n\n\n\n\n# Good news: it can handle unicode characters! \n\ndf9000.loc[df9000.Name.s_string_contains('\u00f6')].Name\n\nOut[14]: \n\n0 Braund, Mr. Owen Harris\u00f6\n\n891 Braund, Mr. Owen Harris\u00f6\n\n1782 Braund, Mr. Owen Harris\u00f6\n\n2673 Braund, Mr. Owen Harris\u00f6\n\n3564 Braund, Mr. Owen Harris\u00f6\n\n4455 Braund, Mr. Owen Harris\u00f6\n\n5346 Braund, Mr. Owen Harris\u00f6\n\n6237 Braund, Mr. Owen Harris\u00f6\n\n7128 Braund, Mr. Owen Harris\u00f6\n\n8019 Braund, Mr. Owen Harris\u00f6\n\nName: Name, dtype: string\n\n\n\n\n\n# Bad news: every time you modify a Series, you have to update it: \n\n\n\ndf9000.loc[df9000.Name.s_string_contains('\u00f6')].Name\n\n0 Braund, Mr. Owen Harris\u00f6\n\n891 Braund, Mr. Owen Harris\u00f6\n\n1782 Braund, Mr. Owen Harris\u00f6\n\n2673 Braund, Mr. Owen Harris\u00f6\n\n3564 Braund, Mr. Owen Harris\u00f6\n\n\n\n\n\ndf9000.loc[df9000.Name.s_string_contains('\u00f6'), \"Name\"] = df9000.loc[df9000.Name.s_string_contains('\u00f6'), \"Name\"] + '\u00c4' # updating \n\n\n\ndf9000.Name\n\n0 Braund, Mr. Owen Harris\u00f6\u00c4\n\n1 Cumings, Mrs. John Bradley (Florence Briggs Th...\n\n2 Heikkinen, Miss. Laina\n\n\n\ndf9000.loc[df9000.Name.s_string_contains('\u00f6'), \"Name\"] # Exception because ds_update_fast_string was not called\n\n\n\nTraceback (most recent call last):\n\n File \"C:\\Users\\Gamer\\anaconda3\\envs\\dfdir\\lib\\site-packages\\IPython\\core\\interactiveshell.py\", line 3398, in run_code\n\n exec(code_obj, self.user_global_ns, self.user_ns)\n\n File \"<ipython-input-7-2b0dfaf8b41c>\", line 1, in <cell line: 1>\n\n df9000.loc[df9000.Name.s_string_contains('\u00f6'), \"Name\"]\n\n File \"C:/Users/Gamer/anaconda3/envs/dfdir/a_pandas_string_search.py\", line 133, in search_contains\n\n wordtosearchbin, columntosearch = _get_col_word(\n\n File \"C:/Users/Gamer/anaconda3/envs/dfdir/a_pandas_string_search.py\", line 103, in _get_col_word\n\n return wordtosearchbin, series._stringser.__array__()\n\nAttributeError: 'NoneType' object has no attribute '__array__'\n\n\n\ndf9000.Name.ds_update_fast_string() # Necessary after changing a Series\n\n# you can also update the whole DataFrame: df9000 = df9000.ds_update_fast_string()\n\n# Be careful: df9000.Name.ds_update_fast_string() returns None (inplace) \n\n# df9000.ds_update_fast_string() returns a DataFrame\n\n\n\ndf9000.loc[df9000.Name.s_string_contains('\u00f6'), \"Name\"] # Now it is working!\n\n\n\n0 Braund, Mr. Owen Harris\u00f6\u00c4\n\n891 Braund, Mr. Owen Harris\u00f6\u00c4\n\n1782 Braund, Mr. Owen Harris\u00f6\u00c4\n\n2673 Braund, Mr. Owen Harris\u00f6\u00c4\n\n```\n\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Up to 4x faster than Series.str.contains / Series.eq - can handle Unicode!",
"version": "0.11",
"split_keywords": [
"pandas",
"dataframe",
"string",
"series",
"fast",
"unicode"
],
"urls": [
{
"comment_text": "",
"digests": {
"md5": "d594debfc4f36af74b726d632da2d42f",
"sha256": "54ad91004bbb9ee0158a77d851ee6fd9db30c2d0d539a92ff452ef1a458b40da"
},
"downloads": -1,
"filename": "a_pandas_ex_fast_string-0.11-py3-none-any.whl",
"has_sig": false,
"md5_digest": "d594debfc4f36af74b726d632da2d42f",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 7603,
"upload_time": "2022-12-20T04:55:22",
"upload_time_iso_8601": "2022-12-20T04:55:22.248088Z",
"url": "https://files.pythonhosted.org/packages/b3/0a/d428ec6ae3cc6b1ff9a6c0d670aeb863cc59f72f4528961183556f8ccd96/a_pandas_ex_fast_string-0.11-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"md5": "3bfb4c18267ed7d985e5e1b99b97e817",
"sha256": "a4a19c7fd7b0f4fc37ddf37ecdd638e5f84f8634471ce32ed91d1114b50ca3ef"
},
"downloads": -1,
"filename": "a_pandas_ex_fast_string-0.11.tar.gz",
"has_sig": false,
"md5_digest": "3bfb4c18267ed7d985e5e1b99b97e817",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 5971,
"upload_time": "2022-12-20T04:55:23",
"upload_time_iso_8601": "2022-12-20T04:55:23.762780Z",
"url": "https://files.pythonhosted.org/packages/05/74/0e7a4707e95520e37bd4f1e3ba8eceadcdd2a7ce78ceb25d633641cea9ff/a_pandas_ex_fast_string-0.11.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2022-12-20 04:55:23",
"github": true,
"gitlab": false,
"bitbucket": false,
"github_user": "hansalemaos",
"github_project": "a_pandas_ex_fast_string",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"requirements": [
{
"name": "numexpr",
"specs": []
},
{
"name": "pandas",
"specs": []
},
{
"name": "ujson",
"specs": []
}
],
"lcname": "a-pandas-ex-fast-string"
}
JSON
