Trace element output format problem

chenhongjun

When I output processed trace elements recently, I found that the value of each element is followed by the error and the detection limit, but we generally want these three values to be divided into three parts when drawing, rather than together. For example, the first part is all element values, the second part is the error, and the third part is the detection limit, so they are displayed separately. I don’t know how to set it when iolite is output. 😀

Bence

Hi chenhongjun,

Those are only combined in the Results Data Table. If you export the data via the Export tab (export to file) the values are separated into different columns so that you can plot them.

Is this what you mean? Am I answering your question correctly?

-Bence

Chengkaide

Hello hongjun
I think you can write a export file, like PlasmaAge.py.

chenhongjun

Chengkaide Thank you very much, I am not very proficient in python, I don’t know how to write a script, if you can, you can teach me

Chengkaide

Bence Hello
My question also need your help.
Please.

chenhongjun

Bence Bence, thank you very much for your prompt answers and help 😃 ，Maybe due to the language, my statement may not be very clear 😃 ，I have currently export the data to an excel file. Although each value in the file is in a different column, the format I want is to output all the values one by one, and then output the error below values, and then output the LODs one by one with a few blank lines below the LODs is similar to the one published in the article, I don’t know if you can understand it, or I’ll cut a picture for you

Bence

chenhongjun
If you could post a picture, that would help me 😃

chenhongjun

Bence https://imgur.com/a/pEugUbh Hello Bence, this link is a screenshot , is it possible to output a similar format? 😃

Bence

Hi chenhongjun

Yes, it is definitely possible, but it would require a custom export script. This is a good example for the iolite python course, if you wanted to write your own?

Or we could write it for you, but it would take a little while as we're running the python course from next week. Whichever works best for you...

chenhongjun

Bence 😅 😄 Dear Bence，Maybe I don’t know how to write this script, thank you for your patience to answer it, and please help me see it when you’re done. :

chenhongjun

Bence 😀 Hi，Bence，I don't know if you are free and can help me to write a script at present :

Bence

Hi @chenhongjun ,

Yes, I can help you write the script. I'll make a start on it and contact you via email. I'll post the result here too, when it's done.

-Bence

chenhongjun

Bence Thank you very much, you are very enthusiastic and help me answer a lot of questions 😄

Bence

Sorry for the delay in getting this to you, but here is a script that you can run to export data in the format you wanted (mean; 1sd; LOD).

You can copy and paste the following into iolite's Python Workspace and click Save As... to save it as a script. You can also call it from the Export View -> Export Script.

Please test it and let me know if you find any problems 😊

-Bence

import pandas as pd
from iolite.QtGui import QFileDialog
from iolite.QtCore import QDir
import sys

def getValuesDF(mode):
    """
    Returns a pandas data frame for all selections in all groups (except
    Baselines).

    The value for each selection to be returned is based on mode:
    Mean: mean of selection
    OneSD: uncertainty as 1 SD
    LOD: LOD (currently just Longerich value)

    Returns pandas dataframe with columns as channels, and each row as a
    selection

    NOTE: The headers are un-named in the returned dataframe. This makes it
    easier to concat/append dataframes
    """
    df = pd.DataFrame()

    selNames = []

    for group in data.selectionGroupList():
        if group.type == data.Baseline:
            continue

        for sel in group.selections():
            if mode == 'Mean':
                res = [data.result(sel, ch).value() for ch in data.timeSeriesList(data.Output)]
            elif mode == 'OneSD':
                res = [data.result(sel, ch).uncertainty("Basic", "AbsOneStdDev") for ch in data.timeSeriesList(data.Output)]
            elif mode == 'LOD':
                res = [data.result(sel, ch).longerichLOD() for ch in data.timeSeriesList(data.Output)]
            else:
                print("ERROR: mode argument must be one of \"Mean\", \"OneSD\", or \"LOD\"")
                return

            df[sel.name] = res
            selNames.append(sel.name)

    # Now transpose it so that each selection is a row instead of a column
    df = df.transpose()

    # And add the sel name as column
    df.insert(0, 'Name', selNames)

    return df


df_vals = getValuesDF('Mean')
df_vals = df_vals.append(pd.Series(), ignore_index=True)

df_unc = getValuesDF('OneSD')
df_unc = df_unc.append(pd.Series(), ignore_index=True)

df_lod = getValuesDF('LOD')

df_final = pd.concat([df_vals, df_unc, df_lod])

# Add the column headings:
headers = [channel.name for channel in data.timeSeriesList(data.Output)]
headers.insert(0, 'Selection Name')
df_final.set_axis(headers, axis=1, inplace=True)

fname = QFileDialog.getSaveFileName(None, "Save Export Data", QDir.homePath(), "CSV Files (*.csv")
if fname:
    df_final.to_csv(fname, index=False)
    print(f"Saved to {fname}")

chenhongjun

Bence Thank you very much for taking the time to help me write this script, I tried to export the data according to the method you said, but the software prompts an error, I don't know the reason 😃 https://imgur.com/sCOAe94

Bence

🤔 I may have been wrong about using it in the Export View -> Export Script.

Can you please try it in the Python Workspace?

chenhongjun

Bence 😀Hi Bence，i have tried it in the Python Workspace But it seems that there are some problems, and the file is not exported. https://imgur.com/we1HWXH Sorry, I am not familiar with Python and don't know what causes this situation.

Bence

That's interesting. There must be something about your dataset that I hadn't thought of. Would you mind sending me a copy of the dataset that is causing this issue, and I'll fix it for you?

Many thanks in advance,
Bence

chenhongjun

Bence Thank you very much, Bence, I found that I have this problem with different batches of data. I selected two sets of data and sent them to the iolite support mailbox, please check 😃

chenhongjun

Bence Hi Bence 😃 , do you receive my email?

Bence

Hi chenhongjun ,

Sorry for the delay in updating this. I have found the problem. Here is an updated version. Please let me know if you have any troubles with it:

import pandas as pd
from iolite.QtGui import QFileDialog
from iolite.QtCore import QDir
import sys

def getValuesDF(mode):
    """
    Returns a pandas data frame for all selections in all groups (except
    Baselines).

    The value for each selection to be returned is based on mode:
    Mean: mean of selection
    OneSD: uncertainty as 1 SD
    LOD: LOD (currently just Longerich value)

    Returns pandas dataframe with columns as channels, and each row as a
    selection

    NOTE: The headers are un-named in the returned dataframe. This makes it
    easier to concat/append dataframes
    """
    df = pd.DataFrame()

    selNames = []

    for group in data.selectionGroupList():
        if group.type == data.Baseline:
            continue

        counter = 1
        for sel in group.selections():
            if mode == 'Mean':
                res = [data.result(sel, ch).value() for ch in data.timeSeriesList(data.Output)]
            elif mode == 'OneSD':
                res = [data.result(sel, ch).uncertainty("Basic", "AbsOneStdDev") for ch in data.timeSeriesList(data.Output)]
            elif mode == 'LOD':
                res = [data.result(sel, ch).longerichLOD() for ch in data.timeSeriesList(data.Output)]
            else:
                print("ERROR: mode argument must be one of \"Mean\", \"OneSD\", or \"LOD\"")
                return

            if sel.name in df.columns:
                this_sel_name = sel.name + '_' + str(counter)
                counter += 1
            else:
                this_sel_name = sel.name


            df[this_sel_name] = res
            selNames.append(this_sel_name)

    # Now transpose it so that each selection is a row instead of a column
    df = df.transpose()

    # And add the sel name as column
    df.insert(0, 'Name', selNames)

    return df


df_vals = getValuesDF('Mean')
df_vals = df_vals.append(pd.Series(), ignore_index=True)

df_unc = getValuesDF('OneSD')
df_unc = df_unc.append(pd.Series(), ignore_index=True)

df_lod = getValuesDF('LOD')

df_final = pd.concat([df_vals, df_unc, df_lod])

# Add the column headings:
headers = [channel.name for channel in data.timeSeriesList(data.Output)]
headers.insert(0, 'Selection Name')
df_final.set_axis(headers, axis=1, inplace=True)

fname = QFileDialog.getSaveFileName(None, "Save Export Data", QDir.homePath(), "CSV Files (*.csv")
if fname:
    df_final.to_csv(fname, index=False)
    print(f"Saved to {fname}")

-Bence