Pydantic error at extract_matched_skills

[hdev-mck]

I'm running into this error, at extract_matched_skills. I checked through the chain and when we create _skills_matcher_chain we defined inputs but there was not parameters passed to the chain. self.llm_kwargs is always [].
Let me know how to mitigate this issue. Thanks

Exception has occurred: ValidationError
3 validation errors for PromptInput
ats_keywords
field required (type=value_error.missing)
non_technical_skills
field required (type=value_error.missing)
technical_skills
field required (type=value_error.missing)
File "ResumeGPT/ResumeGPT/services/resume_improver.py", line 459, in extract_matched_skills
chain_inputs = format_prompt_inputs_as_strings(prompt_inputs=chain.input_schema().dict(), **self.parsed_job, degrees=self.degrees, experiences=utils.dict_to_yaml_string(dict(Experiences=self.experiences)), education=utils.dict_to_yaml_string(dict(Education=self.education)), skills=self._format_skills_for_prompt(self.skills), objective=self.objective)
File "ResumeGPT/ResumeGPT/services/resume_improver.py", line 192, in create_draft_tailored_resume
self.skills = self.extract_matched_skills(verbose=False)
File "ResumeGPT/main.py", line 7, in
resume_improver.create_draft_tailored_resume()
pydantic.v1.error_wrappers.ValidationError: 3 validation errors for PromptInput
ats_keywords
field required (type=value_error.missing)
non_technical_skills
field required (type=value_error.missing)
technical_skills
field required (type=value_error.missing)

[takline]

I suspect it is a formatting issue with the resume yaml file. I've made some changes in the recent commit that should make this a little easier to work with and understand formatting requirements.

If that's not the issue, do you mind sending the link?

[hdev-mck]

I'm still getting the same error. For consistency, what is the python version? you use.

[takline]

3.11.9

From what I can tell, resume_improver.download_and_parse_job_post was unable to extract these required fields:

• ats_keywords
• non_technical_skills
• technical_skills

on this line. Take a look at the output job.yaml file to see if the extractor did in fact fail to extract these fields.

[hdev-mck]

job.yaml is extracting that information.

I think this is a langchain problem. Issue is when we have a chain and try to get input_schema of the chain, it is throwing pydantic errors. I used a test to replicate this issue. Not sure if I'm missing something.

[hdev-mck]

After a bit of grinding, there seems to be pydantic issues with langchain. Some dependency cluster f*. I downgraded to a working version from my previous project and it works.

Below is my requirements.txt file:

`#python 3.11.9

beautifulsoup4
configparser
langchain==0.1.20
langchain-openai==0.1.6
langchain-core
pydantic==2.7.1
reportlab
requests
ruamel.yaml
pytest
free-proxy
langchain-community==0.0.38
python-dateutil
Jinja2
notebook`

[TaherPatan]

It is giving me the same error even with these versions installed in a completely new environment. It is a massive issue.

[takline]

Ah wow appreciate the pointer - I'll make sure to update the requirements to prevent this from happening again

[TaherPatan]

It is giving me the same error even with these versions installed in a completely new environment. It is a massive issue. Did you manage to find any solutions for it apart from installing the packages with the same version?

[virajnp]

Does this tool work and where should the starting point be to run it.

[alwaysLearnNewThings]

I don't see any main.py. do we need to create one?