This page was generated from docs/examples/mineralML_microanalysisoutput.ipynb. Interactive online version: Binder badge.

Python Notebook Download 

[1]:

""" Created on March 26, 2026 // @author: Sarah Shi """

import os
import numpy as np
import pandas as pd

import mineralML as mm
import Thermobar as pt

import matplotlib.pyplot as plt
%matplotlib inline
%config InlineBackend.figure_format = 'png'

mineralML Quickstart for Tabular Microanalytical Data

This notebook shows how to load and run your microanalytical data through mineralML, across a number of different formats returned straight from Cameca, Probe for EPMA, and AZtec (Oxford Instruments). This is a five step process:

  1. Extract and standardize data from raw instrument files (Excel or CSV) using mm.extract_cameca, mm.extract_probe4epma, or mm.extract_aztec. For reference, the common outputs for these three types of instruments is provided on GitHub (https://github.com/sarahshi/mineralML/tree/main/docs/TabularData/microanalysis.xlsx), in the three sheets. Clean and align columns with mm.prep_df.

  2. Run data through the neural network with mm.predict_class_prob to derive classifications and prediction scores.

  3. Export predictions and prediction scores with mm.export_predictions_to_excel.

  4. Calculate mineral compositions and plot compositions in empirical classification space (ternaries, quadrilaterals).

  5. Run data through Thermobar for thermobarometric estimates.

We loaded in the mineralML Python package as mm. mineralML has trained machine learning models for classifying minerals. This implementation aims to get your electron microprobe or quantitative EDS compositions classified and processed. We remove some degrees of freedom to simplify the process as much as possible. The minerals considered for this study include: Amphibole, Apatite, Biotite, Calcite, Chlorite, Epidote, Feldspar (Alkali Feldspar and Plagioclase), Garnet, Glass, Kalsilite, Leucite, Melilite, Muscovite, Nepheline, Olivine, Oxide (Rhombohedral_Oxides including Hematite-Ilmenite, Spinel_Group including Magnetite-Spinel), Pyroxene (Clinopyroxene, Orthopyroxene, Na-Pyroxene), Quartz, Rutile, Serpentine, Titanite, Tourmaline, and Zircon.

An Excel or CSV file containing your electron microprobe or EDS analyses is necessary. Thanks to the integrated extraction functions, you no longer need to manually format complex headers or convert relative errors! Just point the correct extractor to your raw instrument export. The necessary oxides are SiO\(_2\), TiO\(_2\), Al\(_2\)O\(_3\), FeO\(_t\), MnO, MgO, CaO, Na\(_2\)O, K\(_2\)O, Cr\(_2\)O\(_3\), P\(_2\)O\(_5\), and ZrO\(_2\) (if you are aiming to classify zircon). For the oxides not analyzed for specific minerals, the preprocessing will fill in the nan values as 0.

We will apply the neural network method to the dataset.

1. Load and prepare data for analysis

We will use mm.extract_cameca, mm.extract_probe4epma, and mm.extract_aztec to extract the tabular data from their respective sheets in our EPMA dataset. Because these functions standardize the output format, we can easily combine them into a single dataframe before running them through mm.prep_df.

Note that the uncertainties from Cameca EPMAs are provided as 3 sigma! We convert these to 1 sigma.

[2]:

# Define the path to your microanalysis data output. Here, I've written three functions for dealing with Cameca, Probe for EPMA, and AZtec data outputs. If you have other outputs you'd like to work with, reach out and I can expand the functions available.
file_path = 'TabularData/microanalysis.xlsx'

# Extract data from each respective sheet/instrument
df_cameca = mm.extract_cameca(file_path, sheet_name='Cameca')
display('Cameca:', df_cameca.head())

df_p4e = mm.extract_probe4epma(file_path, sheet_name='Probe4EPMA')
display('Probe for EPMA:', df_p4e.head())

df_aztec = mm.extract_aztec(file_path, sheet_name='AZtec')
display('AZtec:', df_aztec.head())

'Cameca:'
Sample SiO2 TiO2 Al2O3 FeOt MnO MgO CaO Na2O K2O ... FeOt_1sigma MnO_1sigma MgO_1sigma CaO_1sigma Na2O_1sigma K2O_1sigma Cr2O3_1sigma P2O5_1sigma NiO_1sigma O_1sigma
0 Museum_Ol174.1_highcount_2 41.77717 0.000017 0.005657 9.425813 0.161341 49.51706 0.017556 0.000013 0.005696 ... 0.077547 0.011184 0.117512 0.006948 -0.000116 0.004680 -0.000028 0.003343 0.015145 NaN
1 Museum_Ol174.1_highcount_2 41.71127 0.000017 0.000019 9.353550 0.121697 49.56913 0.034175 0.003019 0.008749 ... 0.077328 0.011659 0.117604 0.007157 0.004444 0.004610 0.009179 0.003529 0.015307 NaN
2 Museum_Ol174.1_highcount_2 41.74872 0.000017 0.000019 9.297508 0.144290 49.75990 0.011651 0.000013 0.005699 ... 0.076868 0.011469 0.117930 0.006938 -0.000015 0.004808 -0.000013 0.003526 0.015003 NaN
3 Museum_Ol174.1_highcount_2 41.87691 0.000017 0.000019 9.372102 0.116922 49.71903 0.023644 0.000013 0.000012 ... 0.077188 0.012172 0.117876 0.006977 -0.000011 -0.000011 0.009358 0.003245 0.015251 NaN
4 Museum_Ol174.1_highcount_2 41.78258 0.000017 0.003852 9.354717 0.141731 49.81131 0.021923 0.001326 0.003255 ... 0.076843 0.011027 0.118055 0.006793 0.003950 0.004752 0.009304 -0.000064 0.015466 NaN

5 rows × 27 columns

'Probe for EPMA:'
Sample SiO2 TiO2 Al2O3 FeOt MnO MgO CaO Na2O Cr2O3 ... SiO2_1sigma TiO2_1sigma Al2O3_1sigma FeOt_1sigma MnO_1sigma MgO_1sigma CaO_1sigma Na2O_1sigma Cr2O3_1sigma NiO_1sigma
0 TwinLakesDiopside 56.7804 0.0 0.0 0.060834 0.000698 18.4330 25.7335 0.0 0.000000 ... 0.076460 0.0 0.0 0.006379 0.003544 0.027146 0.017425 0.0 0.000000 0.000000
1 TwinLakesDiopside 56.7965 0.0 0.0 0.062401 0.007511 18.4492 25.7467 0.0 0.000000 ... 0.076478 0.0 0.0 0.006488 0.003478 0.027160 0.017426 0.0 0.000000 0.003127
2 TwinLakesDiopside 56.6400 0.0 0.0 0.069621 0.004826 18.4928 25.7743 0.0 0.000000 ... 0.076339 0.0 0.0 0.006467 0.003484 0.027192 0.017432 0.0 0.000000 0.003112
3 TwinLakesDiopside 56.7002 0.0 0.0 0.055659 0.008368 18.4510 25.7420 0.0 0.002227 ... 0.076366 0.0 0.0 0.006336 0.003447 0.027157 0.017423 0.0 0.003274 0.000000
4 TwinLakesDiopside 56.7272 0.0 0.0 0.059733 0.001823 18.4430 25.7682 0.0 0.003180 ... 0.076406 0.0 0.0 0.006341 0.003463 0.027151 0.017429 0.0 0.003251 0.000000

5 rows × 22 columns

'AZtec:'
Sample SiO2 TiO2 Al2O3 FeOt MnO MgO CaO Na2O K2O ... Al2O3_1sigma FeOt_1sigma MnO_1sigma MgO_1sigma CaO_1sigma Na2O_1sigma K2O_1sigma Cr2O3_1sigma SO3_1sigma NiO_1sigma
0 Basalt_113716-1_1_WDS 51.5531 1.3431 15.0794 9.2081 0.1651 8.0553 11.0920 2.7395 0.0719 ... 0.0368 0.0367 0.0154 0.0287 0.0267 0.0227 0.0086 NaN 0.0060 NaN
1 Basalt_113716-1_2_WDS 51.5500 1.3190 15.0192 9.1835 0.1710 8.0690 11.0471 2.7411 0.0699 ... 0.0368 0.0367 0.0154 0.0287 0.0267 0.0227 0.0087 NaN 0.0060 NaN
2 Basalt_113716-1_3_WDS 51.6652 1.3438 15.0686 9.2527 0.1555 8.0666 11.0889 2.7479 0.0775 ... 0.0368 0.0368 0.0154 0.0287 0.0267 0.0227 0.0086 NaN 0.0059 NaN
3 Basalt_113716-1_4_WDS 51.6838 1.3415 15.0678 9.2828 0.2064 8.0310 11.0918 2.7521 0.0809 ... 0.0368 0.0368 0.0154 0.0287 0.0267 0.0227 0.0086 NaN 0.0059 NaN
4 Basalt_113716-1_5_WDS 51.6122 1.3448 15.0311 9.1791 0.1597 8.0274 11.0576 2.7345 0.0740 ... 0.0367 0.0368 0.0154 0.0286 0.0267 0.0227 0.0086 NaN 0.0060 NaN

5 rows × 27 columns

[3]:

# Combine them into a single standardized dataframe
df_load = pd.concat([df_cameca, df_p4e, df_aztec], ignore_index=True)

# Prepare the dataframe by removing rows with too many NaNs, and filling in zeros.
df_nn = mm.prep_df(df_load, # dataframe to prepare
                   renormalize=False, # optionally renormalize rows to sum to 100 wt%
                   convert_fe=False, # optionally convert disparate input formats of Fe all to FeOt
                   drop_empty_rows=False, # optionally drop rows with more nan values than the min_oxide_count
                   min_oxide_count=2, # minimum number of oxides in a row to keep that analysis
                   verbose=True
                   )

# Examine the prepared dataframe
display(df_nn.head())

prep_df: 40 row(s) processed (of 40 input, 0 dropped).

/tmp/ipykernel_3320/1850373241.py:5: UserWarning: The following columns were missing and have been filled with NaN: ['ZrO2', 'Mineral']
  df_nn = mm.prep_df(df_load, # dataframe to prepare
Sample SiO2 TiO2 Al2O3 FeOt MnO MgO CaO Na2O K2O ... Na2O_1sigma K2O_1sigma Cr2O3_1sigma P2O5_1sigma NiO_1sigma O_1sigma O SO3 SO3_1sigma Mineral
0 Museum_Ol174.1_highcount_2 41.77717 0.000017 0.005657 9.425813 0.161341 49.51706 0.017556 0.000013 0.005696 ... -0.000116 0.004680 -0.000028 0.003343 0.015145 NaN NaN NaN NaN NaN
1 Museum_Ol174.1_highcount_2 41.71127 0.000017 0.000019 9.353550 0.121697 49.56913 0.034175 0.003019 0.008749 ... 0.004444 0.004610 0.009179 0.003529 0.015307 NaN NaN NaN NaN NaN
2 Museum_Ol174.1_highcount_2 41.74872 0.000017 0.000019 9.297508 0.144290 49.75990 0.011651 0.000013 0.005699 ... -0.000015 0.004808 -0.000013 0.003526 0.015003 NaN NaN NaN NaN NaN
3 Museum_Ol174.1_highcount_2 41.87691 0.000017 0.000019 9.372102 0.116922 49.71903 0.023644 0.000013 0.000012 ... -0.000011 -0.000011 0.009358 0.003245 0.015251 NaN NaN NaN NaN NaN
4 Museum_Ol174.1_highcount_2 41.78258 0.000017 0.003852 9.354717 0.141731 49.81131 0.021923 0.001326 0.003255 ... 0.003950 0.004752 0.009304 -0.000064 0.015466 NaN NaN NaN NaN NaN

5 rows × 32 columns

2. Apply the trained neural network (mm.predict_class_prob)

We will use mm.predict_class_prob to do so.

[4]:

# The trained neural network can be applied in just one line. It returns predictions in columns called "Predict_Mineral", "Submineral" (if applicable, for pyroxenes, feldspars, and oxides), "Predict_Probability", "Second_Predict_Mineral", "Second_Predict_Probability".

df_pred_nn = mm.predict_class_prob(df_nn)

mineralML: 40 rows — 40 classified by neural network, 0 by empirical rules (Zircon: 0, SiO2 polymorph: 0, Carbonate: 0), 0 skipped (invalid/empty)

[5]:

# Examine the predicted mineral classifications

display(df_pred_nn.head())
Sample SiO2 TiO2 Al2O3 FeOt MnO MgO CaO Na2O K2O Cr2O3 P2O5 ZrO2 Mineral Predict_Mineral Prediction_Score Prediction_Score_Sigma Second_Predict_Mineral Second_Prediction_Score Submineral
0 Museum_Ol174.1_highcount_2 41.77717 0.000017 0.005657 9.425813 0.161341 49.51706 0.017556 0.000013 0.005696 0.000015 0.005393 0.0 NaN Olivine 0.996502 0.052845 Nepheline 0.003324 NaN
1 Museum_Ol174.1_highcount_2 41.71127 0.000017 0.000019 9.353550 0.121697 49.56913 0.034175 0.003019 0.008749 0.009664 0.001659 0.0 NaN Olivine 0.999182 0.007934 Serpentine 0.000765 NaN
2 Museum_Ol174.1_highcount_2 41.74872 0.000017 0.000019 9.297508 0.144290 49.75990 0.011651 0.000013 0.005699 0.000015 0.001245 0.0 NaN Olivine 0.999268 0.007990 Serpentine 0.000495 NaN
3 Museum_Ol174.1_highcount_2 41.87691 0.000017 0.000019 9.372102 0.116922 49.71903 0.023644 0.000013 0.000012 0.016916 0.002075 0.0 NaN Olivine 0.992237 0.072574 Apatite 0.004258 NaN
4 Museum_Ol174.1_highcount_2 41.78258 0.000017 0.003852 9.354717 0.141731 49.81131 0.021923 0.001326 0.003255 0.022817 0.000023 0.0 NaN Olivine 0.999661 0.003711 Serpentine 0.000211 NaN

There is a good amount of information in this dataframe. The predicted mineral is provided in the Predict_Mineral column, along with the prediction score expressed in the Prediction_Score column (representing likelihood of prediction) and standard deviation on this prediction in the Prediction_Score_Sigma column.

[6]:

# Examine the unique predicted minerals.

print(np.unique(df_pred_nn.Predict_Mineral))

['Clinopyroxene' 'Glass' 'Olivine' 'Plagioclase']

3. Export prediction results

Say you would like to go back to working with Excel now. Use mm.export_predictions_to_excel to export the predictions and these values. All the original input data are returned in the first sheet, and data are split into individual mineral phases in all other sheets.

[7]:

# Export prediction results to an Excel workbook with one sheet called "All" containing all rows, and additional sheets for each predicted mineral.

mm.export_predictions_to_excel(df_pred_nn, filename='TabularData/probe_prediction_results.xlsx')

[7]:

'TabularData/probe_prediction_results.xlsx'

4. Calculate mineral components, plot minerals in empirical classification space

We have plagioclase feldspars, olivines, pyroxenes, and glasses here. Let’s calculate olivine forsterite contents, plot up some pyroxene compositions in quadrilateral space, plot up some feldspars in ternary space, and examine some glass compositions in TAS space.

[8]:

# Separate out samples based on their mineralML-derived Predict_Minerals.

fspars = df_pred_nn[df_pred_nn.Predict_Mineral=='Plagioclase']
ols = df_pred_nn[df_pred_nn.Predict_Mineral=='Olivine']
pxs = df_pred_nn[df_pred_nn.Predict_Mineral=='Clinopyroxene']
gls = df_pred_nn[df_pred_nn.Predict_Mineral=='Glass']

  1. Let’s look at the feldspars, and plot them up in component space on the ternary diagram!

[9]:

# Use FeldsparClassifier to examine at the component space (XAn, XAb, XOr)
fspar_comp = mm.FeldsparClassifier(fspars).calculate_components()
display(fspar_comp)

# Use FeldsparClassifier to plot up these data.
fig = mm.FeldsparClassifier(fspars).plot()
Sample SiO2 TiO2 Al2O3 FeOt MnO MgO CaO Na2O K2O ... Prediction_Score Prediction_Score_Sigma Second_Predict_Mineral Second_Prediction_Score Cation_Sum M_site T_site An Ab Or
10 Museum_Albite_highcount_1 69.60268 0.004796 19.76872 0.000013 0.000013 0.005369 0.027527 11.77873 0.013432 ... 0.979612 0.123126 Melilite 0.015786 4.990514 0.986187 4.003825 1.288845e-03 0.997962 0.000749
11 Museum_Albite_highcount_1 69.51132 0.000017 19.69442 0.000013 0.001269 0.005162 0.016962 11.81076 0.008888 ... 0.982219 0.111984 Melilite 0.010359 4.992787 0.989744 4.002662 7.926215e-04 0.998713 0.000495
12 Museum_Albite_highcount_1 69.47325 0.000017 19.76177 0.004375 0.000013 0.001476 0.000014 11.96592 0.015088 ... 0.981024 0.114432 Melilite 0.010888 5.001650 1.000966 3.999164 6.460223e-07 0.999170 0.000829
13 Museum_Albite_highcount_1 69.86915 0.000017 19.74413 0.005984 0.000013 0.005266 0.000014 11.84413 0.021082 ... 0.984747 0.091026 Melilite 0.013292 4.991051 0.987878 4.002217 6.524425e-07 0.998830 0.001170
14 Museum_Albite_highcount_1 69.20001 0.000017 19.58044 0.000013 0.000013 0.004103 0.015394 11.59416 0.015085 ... 0.991223 0.047953 Melilite 0.004294 4.983153 0.977083 4.004729 7.325686e-04 0.998413 0.000855

5 rows × 58 columns

../_images/examples_mineralML_microanalysisoutput_15_1.png

These are albites, and they do indeed plot up in albite space!

  1. Let’s look at the olivines, and calculate up some stoichiometric site occupancies and XFo values!

[10]:

# Use OlivineCalculator to examine stoichiometric/component space (XFo)
ols_comp = mm.OlivineCalculator(ols).calculate_components()
display(ols_comp.head())
Sample SiO2 TiO2 Al2O3 FeOt MnO MgO CaO Na2O K2O ... Predict_Mineral Prediction_Score Prediction_Score_Sigma Second_Predict_Mineral Second_Prediction_Score Cation_Sum M_site T_site M_site_expanded Fo
0 Museum_Ol174.1_highcount_2 41.77717 0.000017 0.005657 9.425813 0.161341 49.51706 0.017556 0.000013 0.005696 ... Olivine 0.996502 0.052845 Nepheline 0.003324 2.989695 1.975342 1.010146 1.979101 0.903515
1 Museum_Ol174.1_highcount_2 41.71127 0.000017 0.000019 9.353550 0.121697 49.56913 0.034175 0.003019 0.008749 ... Olivine 0.999182 0.007934 Serpentine 0.000765 2.990704 1.977335 1.009358 1.980715 0.904275
2 Museum_Ol174.1_highcount_2 41.74872 0.000017 0.000019 9.297508 0.144290 49.75990 0.011651 0.000013 0.005699 ... Olivine 0.999268 0.007990 Serpentine 0.000495 2.991556 1.979606 1.008493 1.982860 0.905124
3 Museum_Ol174.1_highcount_2 41.87691 0.000017 0.000019 9.372102 0.116922 49.71903 0.023644 0.000013 0.000012 ... Olivine 0.992237 0.072574 Apatite 0.004258 2.989806 1.976470 1.009969 1.979469 0.904364
4 Museum_Ol174.1_highcount_2 41.78258 0.000017 0.003852 9.354717 0.141731 49.81131 0.021923 0.001326 0.003255 ... Olivine 0.999661 0.003711 Serpentine 0.000211 2.991909 1.979839 1.007899 1.983302 0.904685

5 rows × 57 columns

All the relevant site information (M, T sites) and forsterite content are calculated and returned!

  1. Let’s look at the pyroxenes, and calculate up stoichiometric site occupancies and examine at the component space (En, Wo, Fs).

[11]:

# Use PyroxeneClassifier to examine stoichiometric/component space (En, Wo, Fs).
pxs_comp = mm.PyroxeneClassifier(pxs).calculate_components()
display(pxs_comp.head())

# Use PyroxeneClassifier to plot up these data.
fig = mm.PyroxeneClassifier(pxs).plot()
Sample SiO2 TiO2 Al2O3 FeOt MnO MgO CaO Na2O K2O ... EnFs DiHd_2003 Di Fe3_Wang21 Fe2_Wang21 Di_h Hd_h Aeg_h Jd_h En_h
15 TwinLakesDiopside 56.7804 0.0 0.0 0.060834 0.000698 18.4330 25.7335 0.0 0.0 ... 0.009220 0.960862 0.959066 -0.039877 0.041687 0.940923 0.0 0.0 0.0 0.019200
16 TwinLakesDiopside 56.7965 0.0 0.0 0.062401 0.007511 18.4492 25.7467 0.0 0.0 ... 0.009243 0.961270 0.959228 -0.039165 0.041020 0.941688 0.0 0.0 0.0 0.019148
17 TwinLakesDiopside 56.6400 0.0 0.0 0.069621 0.004826 18.4928 25.7743 0.0 0.0 ... 0.008481 0.966572 0.964392 -0.033166 0.035239 0.949988 0.0 0.0 0.0 0.016845
18 TwinLakesDiopside 56.7002 0.0 0.0 0.055659 0.008368 18.4510 25.7420 0.0 0.0 ... 0.008763 0.963272 0.961397 -0.037097 0.038754 0.944755 0.0 0.0 0.0 0.018148
19 TwinLakesDiopside 56.7272 0.0 0.0 0.059733 0.001823 18.4430 25.7682 0.0 0.0 ... 0.008185 0.963722 0.961920 -0.037390 0.039168 0.945071 0.0 0.0 0.0 0.017538

5 rows × 86 columns

/home/docs/checkouts/readthedocs.org/user_builds/mineralml/conda/stable/lib/python3.9/site-packages/ternary/plotting.py:148: UserWarning: No data for colormapping provided via 'c'. Parameters 'vmin', 'vmax' will be ignored
  ax.scatter(xs, ys, vmin=vmin, vmax=vmax, **kwargs)
../_images/examples_mineralML_microanalysisoutput_21_2.png

These are diopsides, and they do indeed plot up in diopside space!

  1. Let’s look at the glasses, and calculate up stoichiometric site occupancies and examine them in TAS space.

[12]:

# Use GlassClassifier to return Mg#s and determine the TAS classification.
gls_comp = mm.GlassClassifier(gls).calculate_components()
display(gls_comp)

# Use GlassClassifier to plot up these data.
fig = mm.GlassClassifier(gls).plot()
SiO2 TiO2 Al2O3 FeOt MnO MgO CaO Na2O K2O P2O5 ... K2O_mols MgO_mols MnO_mols Na2O_mols P2O5_mols SiO2_mols TiO2_mols MgNo Mineral TAS
30 51.5531 1.3431 15.0794 9.2081 0.1651 8.0553 11.0920 2.7395 0.0719 0.0 ... 0.000763 0.199862 0.002327 0.044201 0.0 0.858074 0.016817 0.609279 Glass Basalt
31 51.5500 1.3190 15.0192 9.1835 0.1710 8.0690 11.0471 2.7411 0.0699 0.0 ... 0.000742 0.200201 0.002411 0.044226 0.0 0.858023 0.016515 0.610320 Glass Basalt
32 51.6652 1.3438 15.0686 9.2527 0.1555 8.0666 11.0889 2.7479 0.0775 0.0 ... 0.000823 0.200142 0.002192 0.044336 0.0 0.859940 0.016826 0.608462 Glass Basalt
33 51.6838 1.3415 15.0678 9.2828 0.2064 8.0310 11.0918 2.7521 0.0809 0.0 ... 0.000859 0.199259 0.002910 0.044404 0.0 0.860250 0.016797 0.606633 Glass Basalt
34 51.6122 1.3448 15.0311 9.1791 0.1597 8.0274 11.0576 2.7345 0.0740 0.0 ... 0.000786 0.199169 0.002251 0.044120 0.0 0.859058 0.016838 0.609204 Glass Basalt

5 rows × 28 columns

../_images/examples_mineralML_microanalysisoutput_24_1.png

Neat. These are indeed basalts!

5. Perform Thermobarometry

Let’s say that you analyzed these points for thermobarometry, and want to run things through the Thermobar package (Wieser et al., 2022) now. How might you do so? I have written some functions to allow for compatibility with Thermobar, so let’s test things out with some glass analyses as an example.

[13]:

# Filter out the glass from the mineralML prediction in the column, Predict_Mineral. Can also reuse the gls dataframe from above, but remade here for additional reference.
glass = df_pred_nn[df_pred_nn.Predict_Mineral=='Glass']

# Prepare dataframe for thermobarometry by appending the suffix of _Liq to the oxide columns
glass_pt = mm.format_for_thermobar(glass, suffix='_Liq')
display(glass_pt)

# Calculate temperatures with Thermobar, with the Shea et al., 2022 liquid thermometer.
liq_ts = pt.calculate_liq_only_temp(liq_comps=glass_pt,
                                    equationT='T_Shea2022_MgO')
display(liq_ts)
Sample SiO2_Liq TiO2_Liq Al2O3_Liq FeOt_Liq MnO_Liq MgO_Liq CaO_Liq Na2O_Liq K2O_Liq Cr2O3_Liq P2O5_Liq ZrO2_Liq Mineral Predict_Mineral Prediction_Score Prediction_Score_Sigma Second_Predict_Mineral Second_Prediction_Score Submineral
30 Basalt_113716-1_1_WDS 51.5531 1.3431 15.0794 9.2081 0.1651 8.0553 11.0920 2.7395 0.0719 0.0 0.0 0.0 NaN Glass 0.997496 0.008401 Amphibole 0.001572 NaN
31 Basalt_113716-1_2_WDS 51.5500 1.3190 15.0192 9.1835 0.1710 8.0690 11.0471 2.7411 0.0699 0.0 0.0 0.0 NaN Glass 0.997809 0.006814 Amphibole 0.001513 NaN
32 Basalt_113716-1_3_WDS 51.6652 1.3438 15.0686 9.2527 0.1555 8.0666 11.0889 2.7479 0.0775 0.0 0.0 0.0 NaN Glass 0.996068 0.026892 Amphibole 0.003199 NaN
33 Basalt_113716-1_4_WDS 51.6838 1.3415 15.0678 9.2828 0.2064 8.0310 11.0918 2.7521 0.0809 0.0 0.0 0.0 NaN Glass 0.998022 0.006875 Amphibole 0.001234 NaN
34 Basalt_113716-1_5_WDS 51.6122 1.3448 15.0311 9.1791 0.1597 8.0274 11.0576 2.7345 0.0740 0.0 0.0 0.0 NaN Glass 0.998189 0.005033 Amphibole 0.001228 NaN 
30    1460.92236
31    1461.21280
32    1461.16192
33    1460.40720
34    1460.33088
Name: MgO_Liq, dtype: float64

You can do the same thing for any other mineral or melt thermobarometer/chemometer. Append a different suffix for usage with Thermobar! Find one more example of clinopyroxene barometry below.

[14]:

# Filter out the clinopyroxenes from the mineralML prediction in the column, Predict_Mineral. Can also reuse the pxs dataframe from above, but remade here for additional reference.
clinopyroxene = df_pred_nn[df_pred_nn.Predict_Mineral=='Clinopyroxene']

# Prepare dataframe for thermobarometry by appending the suffix of _Cpx to the oxide columns
clinopyroxene_pt = mm.format_for_thermobar(clinopyroxene, suffix='_Cpx')
display(clinopyroxene_pt)

# Calculate temperatures and pressures with Thermobar, with the Putirka, 2008 clinopyroxene thermobarometer.
cpx_ps = pt.calculate_cpx_only_press_temp(cpx_comps=clinopyroxene_pt,
                                          equationT="T_Put2008_eq32d",
                                          equationP="P_Put2008_eq32b",
                                          H2O_Liq=3)
display(cpx_ps)
Sample SiO2_Cpx TiO2_Cpx Al2O3_Cpx FeOt_Cpx MnO_Cpx MgO_Cpx CaO_Cpx Na2O_Cpx K2O_Cpx Cr2O3_Cpx P2O5_Cpx ZrO2_Cpx Mineral Predict_Mineral Prediction_Score Prediction_Score_Sigma Second_Predict_Mineral Second_Prediction_Score Submineral
15 TwinLakesDiopside 56.7804 0.000000 0.000000 0.060834 0.000698 18.4330 25.7335 0.000000 0.0 0.000000 0.0 0.0 NaN Clinopyroxene 0.990018 0.087620 Melilite 0.009753 Wollastonite
16 TwinLakesDiopside 56.7965 0.000000 0.000000 0.062401 0.007511 18.4492 25.7467 0.000000 0.0 0.000000 0.0 0.0 NaN Clinopyroxene 0.995524 0.030773 Melilite 0.004396 Wollastonite
17 TwinLakesDiopside 56.6400 0.000000 0.000000 0.069621 0.004826 18.4928 25.7743 0.000000 0.0 0.000000 0.0 0.0 NaN Clinopyroxene 0.983612 0.109517 Melilite 0.015668 Diopside
18 TwinLakesDiopside 56.7002 0.000000 0.000000 0.055659 0.008368 18.4510 25.7420 0.000000 0.0 0.002227 0.0 0.0 NaN Clinopyroxene 0.997114 0.021070 Melilite 0.002826 Wollastonite
19 TwinLakesDiopside 56.7272 0.000000 0.000000 0.059733 0.001823 18.4430 25.7682 0.000000 0.0 0.003180 0.0 0.0 NaN Clinopyroxene 0.995375 0.025479 Melilite 0.003835 Wollastonite
20 Diopside_117733_1 56.6971 0.014333 0.230047 0.196070 0.033566 18.1725 25.4832 0.190501 0.0 0.003198 0.0 0.0 NaN Clinopyroxene 0.997084 0.016959 Melilite 0.002823 Wollastonite
21 Diopside_117733_1 56.4262 0.013831 0.228832 0.240967 0.033234 18.0318 25.3790 0.203233 0.0 0.000000 0.0 0.0 NaN Clinopyroxene 0.996425 0.029234 Melilite 0.003442 Wollastonite
22 Diopside_117733_1 56.4609 0.013084 0.203597 0.261528 0.030660 18.0912 25.4615 0.173151 0.0 0.000000 0.0 0.0 NaN Clinopyroxene 0.997756 0.010260 Melilite 0.002175 Wollastonite
23 Diopside_117733_1 56.7182 0.009387 0.204781 0.249543 0.029107 18.1683 25.5439 0.164830 0.0 0.003649 0.0 0.0 NaN Clinopyroxene 0.994745 0.034862 Melilite 0.003266 Wollastonite
24 Diopside_117733_1 56.6649 0.009884 0.205739 0.260411 0.029596 18.1841 25.5244 0.170126 0.0 0.000000 0.0 0.0 NaN Clinopyroxene 0.991305 0.064455 Melilite 0.008551 Wollastonite
P_kbar_calc T_K_calc Delta_P_kbar_Iter Delta_T_K_Iter Sample SiO2_Cpx TiO2_Cpx Al2O3_Cpx FeOt_Cpx MnO_Cpx ... Jd_from 0=Na, 1=Al Jd CaTs CaTi DiHd_1996 EnFs DiHd_2003 Di_Cpx FeIII_Wang21 FeII_Wang21
15 NaN NaN NaN NaN NaN 56.7804 0.000000 0.000000 0.060834 0.000698 ... 0 0.000000 0.019891 0.0 0.960956 0.009196 0.960956 0.959160 -0.039783 0.041592
16 NaN NaN NaN NaN NaN 56.7965 0.000000 0.000000 0.062401 0.007511 ... 0 0.000000 0.019535 0.0 0.961364 0.009220 0.961364 0.959322 -0.039070 0.040926
17 NaN NaN NaN NaN NaN 56.6400 0.000000 0.000000 0.069621 0.004826 ... 0 0.000000 0.016536 0.0 0.966666 0.008457 0.966666 0.964486 -0.033072 0.035144
18 NaN NaN NaN NaN NaN 56.7002 0.000000 0.000000 0.055659 0.008368 ... 0 0.000000 0.018470 0.0 0.963367 0.008739 0.963367 0.961492 -0.037003 0.038660
19 NaN NaN NaN NaN NaN 56.7272 0.000000 0.000000 0.059733 0.001823 ... 0 0.000000 0.018603 0.0 0.963816 0.008162 0.963816 0.962015 -0.037296 0.039074
20 -2.682948 1344.722303 0.0 0.0 NaN 56.6971 0.014333 0.230047 0.196070 0.033566 ... 0 0.013144 0.014215 0.0 0.957424 0.006255 0.957424 0.950672 -0.032784 0.038619
21 -2.643540 1336.833754 0.0 0.0 NaN 56.4262 0.013831 0.228832 0.240967 0.033234 ... 0 0.014092 0.013441 0.0 0.959001 0.004758 0.959001 0.950877 -0.032074 0.039280
22 -3.239814 1332.217801 0.0 0.0 NaN 56.4609 0.013084 0.203597 0.261528 0.030660 ... 0 0.011992 0.013467 0.0 0.961054 0.005083 0.961054 0.952413 -0.031057 0.038870
23 -3.176065 1339.951742 0.0 0.0 NaN 56.7182 0.009387 0.204781 0.249543 0.029107 ... 0 0.011369 0.014896 0.0 0.958674 0.006126 0.958674 0.950485 -0.033179 0.040603
24 -3.157676 1339.323094 0.0 0.0 NaN 56.6649 0.009884 0.205739 0.260411 0.029596 ... 0 0.011740 0.013763 0.0 0.959641 0.006486 0.959641 0.951120 -0.031165 0.038917

10 rows × 58 columns

Note that the first 5 analyses do not have any Na2O and thus no jadeite component, so they return pressures of NaNs (cannot get a pressure from this clinopyroxene).

Hopefully, this sort of workflow can facilitate certainty in mineralogy prior to performing thermobarometry!