New model ready for testing!

[github-actions[bot]]

This model is ready for testing. If you are assigned to this issue, please try it out using the CLI, Google Colab and DockerHub and let us know if it works!

[GemmaTuron]

Hi @emmakodes and @febielin Please check that this model works!

[febielin]

I wanted to try testing the model with a random string (not representative of any compound, I assumed) and it still produces an output. Is this unusual behavior? Or, alternatively, is DFD in fact an abbreviated compound that is still recognized by the model? If so, I am curious to learn more about what forms of input the models can recognize and interpret.

FYI I tried this on other random strings and they resulted in errors.

[emmakodes]

Hello @GemmaTuron

CLI

• The model fetch and makes predictions successfully eos9ei3_cli_log.txt eos9ei3_cli_pred.csv

Google Colab

Running Ersilia on google colab is currently not working, when I try to fetch the model, I get this: /bin/bash: ersilia: command not found You can check the notebook here

• Trying to start my Docker Desktop to pull the model image and make predictions

[febielin]

Hi @emmakodes and @GemmaTuron,

I also received the same error on Google Colab when I try fetching the model. /bin/bash: ersilia: command not found

UPDATE Google Colab issue has been resolved. It is functional on colab.

[emmakodes]

Thanks @febielin and @miquelduranfrigola I will test on google colab again

[emmakodes]

DockerHub

• I was able to pull the model image and run prediction:

  root@8262254d6769:~# ersilia -v api run -i "cccc"
  04:53:19 | DEBUG    | Getting session from /root/eos/session.json
  04:53:21 | DEBUG    | Getting session from /root/eos/session.json
  04:53:21 | WARNING  | Lake manager 'isaura' is not installed! We strongly recommend installing it to store calculations persistently
  04:53:21 | ERROR    | Isaura is not installed! Calculations will be done without storing and reading from the lake, unfortunately.
  04:53:45 | DEBUG    | Is fetched: True
  04:53:45 | DEBUG    | Schema available in /root/eos/dest/eos9ei3/api_schema.json
  04:53:45 | DEBUG    | Setting AutoService for eos9ei3
  04:53:45 | INFO     | Service class provided
  04:55:14 | DEBUG    | No file splitting necessary!
  04:55:22 | DEBUG    | Reading card from eos9ei3
  04:55:22 | DEBUG    | Reading shape from eos9ei3
  04:55:22 | DEBUG    | Input Shape: Single
  04:55:22 | DEBUG    | Input type is: compound
  04:55:22 | DEBUG    | Input shape is: Single
  04:55:22 | DEBUG    | Importing module: .types.compound
  04:55:22 | DEBUG    | Checking RDKIT and other requirements necessary for compound inputs
  04:56:14 | DEBUG    | InputShapeSingle shape: Single
  04:56:14 | DEBUG    | API eos9ei3:run initialized at URL http://127.0.0.1:3000
  04:56:14 | DEBUG    | Schema available in /root/eos/dest/eos9ei3/api_schema.json
  04:56:14 | DEBUG    | No file splitting necessary!
  04:56:19 | DEBUG    | Reading card from eos9ei3
  04:56:19 | DEBUG    | Reading shape from eos9ei3
  04:56:19 | DEBUG    | Input Shape: Single
  04:56:19 | DEBUG    | Input type is: compound
  04:56:19 | DEBUG    | Input shape is: Single
  04:56:19 | DEBUG    | Importing module: .types.compound
  04:56:19 | DEBUG    | Checking RDKIT and other requirements necessary for compound inputs
  04:56:19 | DEBUG    | InputShapeSingle shape: Single
  04:56:19 | DEBUG    | API eos9ei3:run initialized at URL http://127.0.0.1:3000
  04:56:19 | DEBUG    | Schema available in /root/eos/dest/eos9ei3/api_schema.json
  04:56:19 | DEBUG    | Posting to run
  04:56:19 | DEBUG    | Batch size 100
  04:56:33 | DEBUG    | Schema available in /root/eos/dest/eos9ei3/api_schema.json
  04:57:52 | DEBUG    | Status code: 200
  04:57:53 | DEBUG    | Schema available in /root/eos/dest/eos9ei3/api_schema.json
  04:57:53 | DEBUG    | Done with unique posting
  {
  "input": {
      "key": "KAKZBPTYRLMSJV-UHFFFAOYSA-N",
      "input": "C=CC=C",
      "text": "C=CC=C"
  },
  "output": {
      "outcome": [
          3.6396
      ]
  }
  }

[emmakodes]

Google Colab

• Model fetch and makes prediction successfully eos9ei3_output.csv

[febielin]

Hello @GemmaTuron @emmakodes,

Tested this model on Google Colab, CLI, and Docker. It is functional.

Documentation:

• CLI: eos9ei3_fetch.txt eos9ei3_output.csv
• Google Colab: https://drive.google.com/file/d/11LIas4cqmazeMLnGf8VdAxac6Kre9Wft/view?usp=sharing
• Docker:

[GemmaTuron]

perfect thanks both, I'll close this issue as completed!