Test model

[GemmaTuron]

Please test that this model works on Google Colab and on Ersilia's CLI

Thanks!

[Zainab-ik]

On it.

[Zainab-ik]

Testing Enviroment - Ersilia CLI

Step 1: Fetch model eos8lok.

Code: ersilia -v fetch eos8lok > eos8lokLogfile.log 2>&1 The model fetched successfully. Attached log file eos8lok_Logfile.log

Step 2: Serve Model

code: ersilia serve eos8lok The model served successfully with Api: run. This is a new API compared with ersilia's APIs which are predict and calculate.

Step 3: Run predictions with model

code: ersilia -v api run -i "CCCC" > eos8lokRun.log 2>&1 The API run is first tested with a single compound CCCC. The model ran successfully output log is attached. eos8lokRun.log

I'd like to test it with the eml_canonical.csv file. code: ersilia -v api run -i eml_canonical.csv -o eos8lokOutput.csv > output.log 2>&1 The eml_canonical.csv was run successfully. Output CSV file attached: eos8lokOutput.csv Output log file: output.log

The model eos8lok was tested and worked successfully on Ersilia CLI

[Zainab-ik]

@GemmaTuron, I'd like to ask how the APIs of the models are generated.

[Zainab-ik]

Testing Environment - Google Colab

Link to code - Colab

The model was successfully fetched, served, and a prediction was made. Attached is the output file - eos8lok_output.csv

Model worked successfully on Google Colab

[GemmaTuron]

Hi @Zainab-ik !

Wonderful, thanks for testing that! Now let's focus on writing your final application

[GemmaTuron]

@AhmedYusuff

Can you also test this model? We always test them twice to make sure they are working in different systems. Thanks!

[GemmaTuron]

@Zainab-ik The model APIs are defined by us, we used to have different ones but now we have decided to call all of them "run" to make it easier

[Zainab-ik]

Hi @Zainab-ik !

Wonderful, thanks for testing that! Now let's focus on writing your final application

Thank you @GemmaTuron

[Zainab-ik]

@Zainab-ik The model APIs are defined by us, we used to have different ones but now we have decided to call all of them "run" to make it easier

That's interesting. So there's no API as predict or calculate any longer. I think it's a great improvement. The results would then determine whether is a calculation or a description, right? @GemmaTuron

[AhmedYusuff]

Hi @GemmaTuron.

EOS8LOK MODEL TEST

I Tested the Model on my Ubuntu 22.04 system and Google Colab.

Result

---

• [x] Model was fetched, served, and Predicted successfully on Google Colab.

• Link to Colab Notebook. Colab

• Colab output file. eos8lok_output.csv

• [x] Model was fetched, served, and Predicted successfully on Ubuntu 22.04.

• Fetch Log. eosfetch.log

• Output File. eos8lokoutput.csv