From 0c08901f07beaa48fcbc94a24b14d9cad890aaf7 Mon Sep 17 00:00:00 2001 From: miguel5612 Date: Wed, 3 Jul 2019 20:36:06 -0500 Subject: [PATCH] MQ303A Saved --- .../MQ303_Regression-checkpoint.ipynb | 1516 +++++++---------- Experiments/MQ303_Regression.ipynb | 1376 +++++++-------- 2 files changed, 1209 insertions(+), 1683 deletions(-) diff --git a/Experiments/.ipynb_checkpoints/MQ303_Regression-checkpoint.ipynb b/Experiments/.ipynb_checkpoints/MQ303_Regression-checkpoint.ipynb index 69d3517..8791c7d 100644 --- a/Experiments/.ipynb_checkpoints/MQ303_Regression-checkpoint.ipynb +++ b/Experiments/.ipynb_checkpoints/MQ303_Regression-checkpoint.ipynb @@ -60,7 +60,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 4, "metadata": {}, "outputs": [ { @@ -98,20 +98,19 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 5, "metadata": {}, "outputs": [], "source": [ - "x_MQ303A = sheetMQ3.col_values(0)[2:]\n", - "MQ5_CH4 = sheetMQ3.col_values(1)[2:]\n", - "MQ5_CO = sheetMQ3.col_values(2)[2:]\n", - "MQ5_H2 = sheetMQ3.col_values(3)[2:]\n", - "MQ5_Alcohol = sheetMQ3.col_values(4)[2:]" + "x_MQ303A = sheetMQ303A.col_values(0)[2:]\n", + "MQ303A_Iso_butano = sheetMQ303A.col_values(1)[2:]\n", + "MQ303A_Hidrogeno = sheetMQ303A.col_values(2)[2:]\n", + "MQ303A_Alcohol = sheetMQ303A.col_values(3)[2:]" ] }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 7, "metadata": {}, "outputs": [], "source": [ @@ -122,19 +121,18 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 8, "metadata": {}, "outputs": [], "source": [ - "MQ5_CH4 =zero_to_nan(MQ5_CH4)\n", - "MQ5_CO =zero_to_nan(MQ5_CO)\n", - "MQ5_H2 =zero_to_nan(MQ5_H2)\n", - "MQ5_Alcohol =zero_to_nan(MQ5_Alcohol)" + "MQ303A_Iso_butano =zero_to_nan(MQ303A_Iso_butano)\n", + "MQ303A_Hidrogeno =zero_to_nan(MQ303A_Hidrogeno)\n", + "MQ303A_Alcohol =zero_to_nan(MQ303A_Alcohol)" ] }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 12, "metadata": { "scrolled": false }, @@ -148,109 +146,86 @@ "from sklearn import datasets\n", "from sklearn import linear_model\n", "\n", - "dataCH4 = {'RsRo': x_MQ309, 'CH4': MQ5_CH4}\n", - "dataCO = {'RsRo': x_MQ309, 'CO': MQ5_CO}\n", - "dataH2 = {'RsRo': x_MQ309, 'H2': MQ5_H2}\n", - "dataLPG = {'RsRo': x_MQ309, 'Alcohol': MQ5_Alcohol}\n", + "dataIso_Butano = {'RsRo': x_MQ303A, 'Iso_Butano': MQ303A_Iso_butano}\n", + "dataHidrogeno = {'RsRo': x_MQ303A, 'Hidrogeno': MQ303A_Hidrogeno}\n", + "dataAlcohol = {'RsRo': x_MQ303A, 'Alcohol': MQ303A_Alcohol}\n", "\n", - "dfMQ309A_CH4 = pd.DataFrame(dataCH4)\n", - "dfMQ309A_CO = pd.DataFrame(dataCO)\n", - "dfMQ309A_H2 = pd.DataFrame(dataH2)\n", - "dfMQ309A_Alcohol = pd.DataFrame(dataLPG)\n", + "dfMQ303A_Iso_Butano = pd.DataFrame(dataIso_Butano)\n", + "dfMQ303A_Hidrogeno = pd.DataFrame(dataHidrogeno)\n", + "dfMQ303A_Alcohol = pd.DataFrame(dataAlcohol)\n", "\n", - "dfMQ309A_CH4['CH4'] = pd.to_numeric(dfMQ309A_CH4['CH4'])\n", - "dfMQ309A_CO['CO'] = pd.to_numeric(dfMQ309A_CO['CO'])\n", - "dfMQ309A_H2['H2'] = pd.to_numeric(dfMQ309A_H2['H2'])\n", - "dfMQ309A_Alcohol['Alcohol'] = pd.to_numeric(dfMQ309A_Alcohol['Alcohol'])\n", + "dfMQ303A_Iso_Butano['Iso_Butano'] = pd.to_numeric(dfMQ303A_Iso_Butano['Iso_Butano'])\n", + "dfMQ303A_Hidrogeno['Hidrogeno'] = pd.to_numeric(dfMQ303A_Hidrogeno['Hidrogeno'])\n", + "dfMQ303A_Alcohol['Alchol'] = pd.to_numeric(dfMQ303A_Alcohol['Alcohol'])\n", "\n", - "dfMQ309A_CH4['CH4'] = dfMQ309A_CH4['CH4'].replace('',None, regex=True)\n", - "dfMQ309A_CO['CO'] = dfMQ309A_CO['CO'].replace('',None, regex=True)\n", - "dfMQ309A_H2['H2'] = dfMQ309A_H2['H2'].replace('',None, regex=True)\n", - "dfMQ309A_Alcohol['Alcohol'] = dfMQ309A_Alcohol['Alcohol'].replace('',None, regex=True)\n", + "dfMQ303A_Iso_Butano['Iso_Butano'] = dfMQ303A_Iso_Butano['Iso_Butano'].replace('',None, regex=True)\n", + "dfMQ303A_Hidrogeno['Hidrogeno'] = dfMQ303A_Hidrogeno['Hidrogeno'].replace('',None, regex=True)\n", + "dfMQ303A_Alcohol['Alchol'] = dfMQ303A_Alcohol['Alchol'].replace('',None, regex=True)\n", "\n", "#Global X_Predict variable\n", - "X_Predict = dfMQ309A_CH4.RsRo.apply(lambda x: [x]).tolist()" + "X_Predict = dfMQ303A_Iso_Butano.RsRo.apply(lambda x: [x]).tolist()" ] }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 14, "metadata": {}, "outputs": [], "source": [ - "#Model and train CH4\n", - "dataset2TrainCH4 = dfMQ309A_CH4.copy()\n", - "dataset2TrainCH4.dropna(inplace=True)\n", - "X_trainCH4 = dataset2TrainCH4.RsRo.apply(lambda x: [x]).tolist()\n", - "y_trainCH4 = dataset2TrainCH4['CH4'].tolist()\n", + "#Model and train Iso-Butano\n", + "dataset2TrainIso_Butano = dfMQ303A_Iso_Butano.copy()\n", + "dataset2TrainIso_Butano.dropna(inplace=True)\n", + "X_trainIso_Butano = dataset2TrainIso_Butano.RsRo.apply(lambda x: [x]).tolist()\n", + "y_trainIso_Butano = dataset2TrainIso_Butano['Iso_Butano'].tolist()\n", "model = linear_model.Lasso(alpha=0.1)\n", - "model.fit(X_trainCH4, y_trainCH4)\n", + "model.fit(X_trainIso_Butano, y_trainIso_Butano)\n", "#Predict\n", - "CH4_Predicted = model.predict(X_Predict)\n", + "Iso_Butano_Predicted = model.predict(X_Predict)\n", "#save into MQ2\n", - "MQ309A_CH4 = CH4_Predicted" + "MQ303A_Iso_Butano = Iso_Butano_Predicted" ] }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 16, "metadata": {}, "outputs": [], "source": [ - "#Model and train CO\n", - "dataset2TrainCO = dfMQ309A_CO.copy()\n", - "dataset2TrainCO.dropna(inplace=True)\n", - "X_trainCO = dataset2TrainCO.RsRo.apply(lambda x: [x]).tolist()\n", - "y_trainCO = dataset2TrainCO['CO'].tolist()\n", + "#Model and train Hidrogeno\n", + "dataset2TrainHidrogeno = dfMQ303A_Hidrogeno.copy()\n", + "dataset2TrainHidrogeno.dropna(inplace=True)\n", + "X_trainHidrogeno = dataset2TrainHidrogeno.RsRo.apply(lambda x: [x]).tolist()\n", + "y_trainHidrogeno = dataset2TrainHidrogeno['Hidrogeno'].tolist()\n", "model = linear_model.Lasso(alpha=0.1)\n", - "model.fit(X_trainCO, y_trainCO)\n", + "model.fit(X_trainHidrogeno, y_trainHidrogeno)\n", "#Predict\n", - "CO_Predicted = model.predict(X_Predict)\n", + "Hidrogeno_Predicted = model.predict(X_Predict)\n", "#save into MQ2\n", - "MQ309A_CO = CO_Predicted" + "MQ303A_Hidrogeno = Hidrogeno_Predicted" ] }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 19, "metadata": {}, "outputs": [], "source": [ - "#Model and train H2\n", - "dataset2TrainH2 = dfMQ309A_H2.copy()\n", - "dataset2TrainH2.dropna(inplace=True)\n", - "X_trainH2 = dataset2TrainH2.RsRo.apply(lambda x: [x]).tolist()\n", - "y_trainH2 = dataset2TrainH2['H2'].tolist()\n", + "#Model and train Alchol\n", + "dataset2TrainAlchol = dfMQ303A_Alcohol.copy()\n", + "dataset2TrainAlchol.dropna(inplace=True)\n", + "X_trainAlchol = dataset2TrainAlchol.RsRo.apply(lambda x: [x]).tolist()\n", + "y_trainAlchol = dataset2TrainAlchol['Alchol'].tolist()\n", "model = linear_model.Lasso(alpha=0.1)\n", - "model.fit(X_trainH2, y_trainH2)\n", + "model.fit(X_trainAlchol, y_trainAlchol)\n", "#Predict\n", - "H2_Predicted = model.predict(X_Predict)\n", + "Alchol_Predicted = model.predict(X_Predict)\n", "#save into MQ2\n", - "MQ309A_H2 = H2_Predicted\n" + "MQ303A_Alchol = Alchol_Predicted\n" ] }, { "cell_type": "code", - "execution_count": 10, - "metadata": {}, - "outputs": [], - "source": [ - "#Model and train Alcohol\n", - "dataset2TrainAlcohol = dfMQ309A_Alcohol.copy()\n", - "dataset2TrainAlcohol.dropna(inplace=True)\n", - "X_trainAlcohol = dataset2TrainAlcohol.RsRo.apply(lambda x: [x]).tolist()\n", - "y_trainAlcohol = dataset2TrainAlcohol['Alcohol'].tolist()\n", - "model = linear_model.Lasso(alpha=0.1)\n", - "model.fit(X_trainAlcohol, y_trainAlcohol)\n", - "#Predict\n", - "Alcohol_Predicted = model.predict(X_Predict)\n", - "#save into MQ2\n", - "MQ309A_Alcohol = Alcohol_Predicted" - ] - }, - { - "cell_type": "code", - "execution_count": 13, + "execution_count": 21, "metadata": {}, "outputs": [ { @@ -260,7 +235,7 @@ "\r\n", "\r\n", - "\r\n", + "\r\n", " \r\n", " \r\n" @@ -2016,10 +1767,9 @@ "fig.set_dpi(200)\n", "\n", "# only these two lines are calibration curves\n", - "plt.plot(MQ309A_CH4, x_MQ309A, marker='o', linewidth=1, label='CH4')\n", - "plt.plot(MQ309A_CO, x_MQ309A, marker='o', linewidth=1, label='CO')\n", - "plt.plot(MQ309A_H2, x_MQ309A, marker='o', linewidth=1, label='H2')\n", - "plt.plot(MQ309A_Alcohol, x_MQ309A, marker='o', linewidth=1, label='Alcohol')\n", + "plt.plot(MQ303A_Iso_Butano, x_MQ303A, marker='o', linewidth=1, label='Iso-Butano')\n", + "plt.plot(MQ303A_Hidrogeno, x_MQ303A, marker='o', linewidth=1, label='Hidrogeno')\n", + "plt.plot(MQ303A_Alchol, x_MQ303A, marker='o', linewidth=1, label='Alcohol')\n", "\n", "\n", "\n", @@ -2034,15 +1784,15 @@ "\n", "plt.grid(b=True, which='minor', color='lightgrey', linestyle='--')\n", "\n", - "fig.suptitle('Calibration plot for MQ-309A data')\n", + "fig.suptitle('Calibration plot for MQ-303A data')\n", "ax.set_xlabel('PPM Concentration')\n", "ax.set_ylabel('Rs/Ro')\n", "\n", "\n", "#Save image\n", - "plt.savefig('MQ309A.svg', format = 'svg', dpi = 1200)\n", - "plt.savefig('MQ309A.png')\n", - "plt.savefig('MQ309A.eps', format = 'eps', dpi = 1200)" + "plt.savefig('MQ303A.svg', format = 'svg', dpi = 1200)\n", + "plt.savefig('MQ303A.png')\n", + "plt.savefig('MQ303A.eps', format = 'eps', dpi = 1200)" ] }, { diff --git a/Experiments/MQ303_Regression.ipynb b/Experiments/MQ303_Regression.ipynb index 90a2948..8791c7d 100644 --- a/Experiments/MQ303_Regression.ipynb +++ b/Experiments/MQ303_Regression.ipynb @@ -225,7 +225,7 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 21, "metadata": {}, "outputs": [ { @@ -235,7 +235,7 @@ "\r\n", "\r\n", - "\r\n", + "\r\n", " \r\n", " \r\n" @@ -1992,8 +1768,8 @@ "\n", "# only these two lines are calibration curves\n", "plt.plot(MQ303A_Iso_Butano, x_MQ303A, marker='o', linewidth=1, label='Iso-Butano')\n", - "plt.plot(MQ303A_CO, x_MQ303A, marker='o', linewidth=1, label='CO')\n", - "plt.plot(MQ303A_Alcohol, x_MQ303A, marker='o', linewidth=1, label='Alcohol')\n", + "plt.plot(MQ303A_Hidrogeno, x_MQ303A, marker='o', linewidth=1, label='Hidrogeno')\n", + "plt.plot(MQ303A_Alchol, x_MQ303A, marker='o', linewidth=1, label='Alcohol')\n", "\n", "\n", "\n",