{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Introduction\n", "\n", "This document is intended to describe how to access data from a MySQL database using Python. It utilizes a database of wideband acoustic immitance variables from humans with normal hearing (see https://projectreporter.nih.gov/project_info_description.cfm?aid=8769352&icde=30039221&ddparam=&ddvalue=&ddsub=&cr=10&csb=default&cs=ASC for more details).\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Accessing data from a database using SQL commands\n", "\n", "First I will demonstrate how to access data using SQL (structured query language) commands and the `mysql.connector.connect()` function. We begin by setting up a connection \n", "to the database." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Connection to MySQL DB successful\n", "\n" ] } ], "source": [ "import mysql.connector\n", "import pandas as pd\n", "import seaborn as sns\n", "import matplotlib.pyplot as plt\n", "\n", "%matplotlib inline\n", "\n", "\n", "# A function that can create a connection to a database\n", "def create_connection(host_name, user_name, user_password, db_name):\n", " connection = None\n", " try:\n", " connection = mysql.connector.connect(\n", " host=host_name,\n", " user=user_name,\n", " passwd=user_password,\n", " database=db_name\n", " )\n", " print(\"Connection to MySQL DB successful\")\n", " except Error as e:\n", " print(f\"The error '{e}' occurred\")\n", "\n", " return connection\n", "\n", "\n", "# connect to Nick's databse at Smith\n", "con = create_connection(\"scidb.smith.edu\", \"waiuser\", \"smith_waiDB\", \"wai\")\n", "\n", "print(con)\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Executing SQL queries\n", "\n", "Next a series of SQL queries can be sent to the database. These return lists of tuples which we can convert to a pandas DataFrame using pd.DataFrame(). \n", "\n" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Tables
0Codebook
1Measurements
2Measurements_pre2020
3PI_Info
4PI_Info_OLD
5Subjects
6Subjects_pre2020
\n", "
" ], "text/plain": [ " Tables\n", "0 Codebook\n", "1 Measurements\n", "2 Measurements_pre2020\n", "3 PI_Info\n", "4 PI_Info_OLD\n", "5 Subjects\n", "6 Subjects_pre2020" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
FieldTypeNullKeyDefaultExtra
0Identifiervarchar(50)NOPRINone
1Yearint(11)NONone
2AuthorstextNONone
3AuthorsShortListtextNONone
4TitletextNONone
5JournaltextNONone
6URLtextNONone
7AbstracttextNONone
8DataSubmitterNametextNONone
9DataSubmitterEmailtextNONone
10DateSubmittedtextNONone
11PI_NotestextNONone
\n", "
" ], "text/plain": [ " Field Type Null Key Default Extra\n", "0 Identifier varchar(50) NO PRI None \n", "1 Year int(11) NO None \n", "2 Authors text NO None \n", "3 AuthorsShortList text NO None \n", "4 Title text NO None \n", "5 Journal text NO None \n", "6 URL text NO None \n", "7 Abstract text NO None \n", "8 DataSubmitterName text NO None \n", "9 DataSubmitterEmail text NO None \n", "10 DateSubmitted text NO None \n", "11 PI_Notes text NO None " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
FieldTypeNullKeyDefaultExtra
0Identifiervarchar(50)NOPRINone
1SubjectNumberint(11)NOPRINone
2SessionTotalint(11)NONone
3AgeFirstMeasurementfloatYESNone
4AgeCategoryFirstMeasurementvarchar(50)YESNone
5Sexvarchar(50)NONone
6Racevarchar(50)NONone
7Ethnicityvarchar(50)NONone
8LeftEarStatusFirstMeasurementvarchar(50)NONone
9RightEarStatusFirstMeasurementvarchar(50)NONone
10SubjectNotestextYESNone
\n", "
" ], "text/plain": [ " Field Type Null Key Default Extra\n", "0 Identifier varchar(50) NO PRI None \n", "1 SubjectNumber int(11) NO PRI None \n", "2 SessionTotal int(11) NO None \n", "3 AgeFirstMeasurement float YES None \n", "4 AgeCategoryFirstMeasurement varchar(50) YES None \n", "5 Sex varchar(50) NO None \n", "6 Race varchar(50) NO None \n", "7 Ethnicity varchar(50) NO None \n", "8 LeftEarStatusFirstMeasurement varchar(50) NO None \n", "9 RightEarStatusFirstMeasurement varchar(50) NO None \n", "10 SubjectNotes text YES None " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
FieldTypeNullKeyDefaultExtra
0Identifiervarchar(50)NOPRINone
1SubjectNumberint(11)NOPRINone
2Sessionint(11)NOPRINone
3Earvarchar(50)NOPRI
4Instrumentvarchar(50)NOPRI
5AgefloatYESNone
6AgeCategoryvarchar(50)YESNone
7EarStatusvarchar(50)YESNone
8TPPfloatYESNone
9AreaCanalfloatYESNone
10PressureCanalfloatNOPRI0
11SweepDirectionvarchar(50)NOPRI
12FrequencyfloatNOPRI0
13AbsorbancefloatYESNone
14ZmagfloatYESNone
15ZangfloatYESNone
\n", "
" ], "text/plain": [ " Field Type Null Key Default Extra\n", "0 Identifier varchar(50) NO PRI None \n", "1 SubjectNumber int(11) NO PRI None \n", "2 Session int(11) NO PRI None \n", "3 Ear varchar(50) NO PRI \n", "4 Instrument varchar(50) NO PRI \n", "5 Age float YES None \n", "6 AgeCategory varchar(50) YES None \n", "7 EarStatus varchar(50) YES None \n", "8 TPP float YES None \n", "9 AreaCanal float YES None \n", "10 PressureCanal float NO PRI 0 \n", "11 SweepDirection varchar(50) NO PRI \n", "12 Frequency float NO PRI 0 \n", "13 Absorbance float YES None \n", "14 Zmag float YES None \n", "15 Zang float YES None " ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "# get a list of tables in the database \n", "cursor = con.cursor()\n", "cursor.execute(\"SHOW TABLES\")\n", "display(pd.DataFrame(cursor.fetchall(), columns = [\"Tables\"]))\n", "\n", "\n", "# get information on these tables \n", "explain_col_names = [\"Field\", \"Type\", \"Null\", \"Key\", \"Default\", \"Extra\"] \n", "\n", "cursor.execute(\"EXPLAIN PI_Info\")\n", "PI_Info_table_info = pd.DataFrame(cursor.fetchall(), columns = explain_col_names)\n", "display(PI_Info_table_info)\n", "\n", "cursor.execute(\"EXPLAIN Subjects\")\n", "Subjects_table_info = pd.DataFrame(cursor.fetchall(), columns = explain_col_names)\n", "display(Subjects_table_info)\n", "\n", "cursor.execute(\"EXPLAIN Measurements\")\n", "Measurements_table_info = pd.DataFrame(cursor.fetchall(), columns = explain_col_names)\n", "display(Measurements_table_info)\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Let's explore the `PI_Info` table." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
01234567891011
0Abur_20142014Defne Abur, Nicholas J. Horton, and Susan E. VossAbur et al.Instrasubject variability in power reflectanceJ Am Acad Audiolhttps://www.ncbi.nlm.nih.gov/pubmed/25257718\"<p> <strong> Background: </strong> Power refl...Susan Vosssvoss@smith.edu8/24/16Measurements made on 7 subjects across multipl...
1Feeney_20172017M. Patrick Feeney, Douglas H. Keefe, Lisa L. ...Feeney et al.Normative wideband reflectance, equivalent adm...Ear Hearhttps://www.ncbi.nlm.nih.gov/pubmed/28045835\"<p> <strong> Objectives: </strong> Wideband a...M. Patrick Feeney; Douglas H. KeefePatrick.Feeney@va.gov; Douglas.Keefe@boystown.org6/7/18Database includes measurements on 32 subjects,...
2Groon_20152015Katherine A. Groon, Daniel M. Rasetshwane, Jud...Groon et al.Air-leak effects on ear-canal acoustic absorbanceEar Hearhttps://journals.lww.com/ear-hearing/fulltext/...\"<p> <strong> Objective: </strong> Accurate ea...Steve NeelyStephen.Neely@boystown.org6/18/19Data collected on system described in Rasetshw...
\n", "
" ], "text/plain": [ " 0 1 2 \\\n", "0 Abur_2014 2014 Defne Abur, Nicholas J. Horton, and Susan E. Voss \n", "1 Feeney_2017 2017 M. Patrick Feeney, Douglas H. Keefe, Lisa L. ... \n", "2 Groon_2015 2015 Katherine A. Groon, Daniel M. Rasetshwane, Jud... \n", "\n", " 3 4 \\\n", "0 Abur et al. Instrasubject variability in power reflectance \n", "1 Feeney et al. Normative wideband reflectance, equivalent adm... \n", "2 Groon et al. Air-leak effects on ear-canal acoustic absorbance \n", "\n", " 5 6 \\\n", "0 J Am Acad Audiol https://www.ncbi.nlm.nih.gov/pubmed/25257718 \n", "1 Ear Hear https://www.ncbi.nlm.nih.gov/pubmed/28045835 \n", "2 Ear Hear https://journals.lww.com/ear-hearing/fulltext/... \n", "\n", " 7 \\\n", "0 \"

Background: Power refl... \n", "1 \"

Objectives: Wideband a... \n", "2 \"

Objective: Accurate ea... \n", "\n", " 8 \\\n", "0 Susan Voss \n", "1 M. Patrick Feeney; Douglas H. Keefe \n", "2 Steve Neely \n", "\n", " 9 10 \\\n", "0 svoss@smith.edu 8/24/16 \n", "1 Patrick.Feeney@va.gov; Douglas.Keefe@boystown.org 6/7/18 \n", "2 Stephen.Neely@boystown.org 6/18/19 \n", "\n", " 11 \n", "0 Measurements made on 7 subjects across multipl... \n", "1 Database includes measurements on 32 subjects,... \n", "2 Data collected on system described in Rasetshw... " ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "cursor.execute(\"SELECT * from PI_Info\")\n", "PI_table = pd.DataFrame(cursor.fetchall())\n", "PI_table.head(3)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Let's explore the `Subjects` table." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/html": [ "

\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
FieldIdentifierSubjectNumberSessionTotalAgeFirstMeasurementAgeCategoryFirstMeasurementSexRaceEthnicityLeftEarStatusFirstMeasurementRightEarStatusFirstMeasurementSubjectNotes
0Abur_20141720.0AdultFemaleUnknownUnknownNormalNormal
1Abur_20143819.0AdultFemaleUnknownUnknownNormalNormalSession 5 not included do to acoustic leak
2Abur_20144721.0AdultFemaleUnknownUnknownNormalNormal
\n", "
" ], "text/plain": [ "Field Identifier SubjectNumber SessionTotal AgeFirstMeasurement \\\n", "0 Abur_2014 1 7 20.0 \n", "1 Abur_2014 3 8 19.0 \n", "2 Abur_2014 4 7 21.0 \n", "\n", "Field AgeCategoryFirstMeasurement Sex Race Ethnicity \\\n", "0 Adult Female Unknown Unknown \n", "1 Adult Female Unknown Unknown \n", "2 Adult Female Unknown Unknown \n", "\n", "Field LeftEarStatusFirstMeasurement RightEarStatusFirstMeasurement \\\n", "0 Normal Normal \n", "1 Normal Normal \n", "2 Normal Normal \n", "\n", "Field SubjectNotes \n", "0 \n", "1 Session 5 not included do to acoustic leak \n", "2 " ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "cursor.execute(\"SELECT * from Subjects\")\n", "subject_table = pd.DataFrame(cursor.fetchall(), columns = Subjects_table_info[\"Field\"])\n", "subject_table.head(3)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Let's explore the `Measurements` table.\n", "\n", "Probably best to only get a subset of this table because it is very large!\n", "\n" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
FieldIdentifierSubjectNumberSessionEarInstrumentAgeAgeCategoryEarStatusTPPAreaCanalPressureCanalSweepDirectionFrequencyAbsorbanceZmagZang
0Abur_201411LeftHearID20.0AdultNormal-5.00.0000440.0Ambient210.9380.033338113780000.0-0.233504
1Abur_201411LeftHearID20.0AdultNormal-5.00.0000440.0Ambient234.3750.031571103585000.0-0.235778
2Abur_201411LeftHearID20.0AdultNormal-5.00.0000440.0Ambient257.8120.04057592951700.0-0.233482
\n", "
" ], "text/plain": [ "Field Identifier SubjectNumber Session Ear Instrument Age AgeCategory \\\n", "0 Abur_2014 1 1 Left HearID 20.0 Adult \n", "1 Abur_2014 1 1 Left HearID 20.0 Adult \n", "2 Abur_2014 1 1 Left HearID 20.0 Adult \n", "\n", "Field EarStatus TPP AreaCanal PressureCanal SweepDirection Frequency \\\n", "0 Normal -5.0 0.000044 0.0 Ambient 210.938 \n", "1 Normal -5.0 0.000044 0.0 Ambient 234.375 \n", "2 Normal -5.0 0.000044 0.0 Ambient 257.812 \n", "\n", "Field Absorbance Zmag Zang \n", "0 0.033338 113780000.0 -0.233504 \n", "1 0.031571 103585000.0 -0.235778 \n", "2 0.040575 92951700.0 -0.233482 " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/plain": [ "(626945, 16)" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "cursor.execute(\"SELECT * from Measurements\")\n", "measurements_table = pd.DataFrame(cursor.fetchall(), columns = Measurements_table_info[\"Field\"])\n", "display(measurements_table.head(3))\n", "\n", "measurements_table.shape\n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Let's extract data from a given subject\n" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
FieldIdentifierSubjectNumberSessionEarInstrumentAgeAgeCategoryEarStatusTPPAreaCanalPressureCanalSweepDirectionFrequencyAbsorbanceZmagZang
501097Rosowski_201231LeftHearID30.0AdultNormalNaNNaN0.0Ambient210.9380.08520176591100.0-0.220494
501098Rosowski_201231LeftHearID30.0AdultNormalNaNNaN0.0Ambient234.3750.09034566884300.0-0.222228
501099Rosowski_201231LeftHearID30.0AdultNormalNaNNaN0.0Ambient257.8120.11152758816400.0-0.219561
\n", "
" ], "text/plain": [ "Field Identifier SubjectNumber Session Ear Instrument Age \\\n", "501097 Rosowski_2012 3 1 Left HearID 30.0 \n", "501098 Rosowski_2012 3 1 Left HearID 30.0 \n", "501099 Rosowski_2012 3 1 Left HearID 30.0 \n", "\n", "Field AgeCategory EarStatus TPP AreaCanal PressureCanal SweepDirection \\\n", "501097 Adult Normal NaN NaN 0.0 Ambient \n", "501098 Adult Normal NaN NaN 0.0 Ambient \n", "501099 Adult Normal NaN NaN 0.0 Ambient \n", "\n", "Field Frequency Absorbance Zmag Zang \n", "501097 210.938 0.085201 76591100.0 -0.220494 \n", "501098 234.375 0.090345 66884300.0 -0.222228 \n", "501099 257.812 0.111527 58816400.0 -0.219561 " ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "one_subj = measurements_table[(measurements_table.Identifier == \"Rosowski_2012\") & (measurements_table.SubjectNumber == 3)]\n", "\n", "one_subj.head(3)\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Finally we can plot the results" ] }, { "cell_type": "code", "execution_count": 133, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "splot = sns.relplot(data = one_subj, x = \"Frequency\", y = \"Absorbance\", hue = \"Ear\");\n", "splot.set(xscale=\"log\");\n", "splot.set(title = \"Absorbance by ear Rosowski subject 3\");\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Close the connection to the database" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "con.close()" ] } ], "metadata": { "kernelspec": { "display_name": "Python [conda env:.conda-wav_personal]", "language": "python", "name": "conda-env-.conda-wav_personal-py" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.4" } }, "nbformat": 4, "nbformat_minor": 2 }