CotQAEvalChain

Utilize the LLM generate method for speed gains.

Create outputs from response.

Format prompt with kwargs and pass to LLM.

Format prompt with kwargs and pass to LLM.

Call predict and then parse the results.

Call apredict and then parse the results.

Call apply and then parse the results.

Call apply and then parse the results.

Create LLMChain from LLM and template.

Asynchronously execute the chain.

Validate and prepare chain outputs, and save info about this run to memory.

Validate and prepare chain outputs, and save info about this run to memory.

Prepare chain inputs, including adding inputs from memory.

Prepare chain inputs, including adding inputs from memory.

Convenience method for executing chain.

Convenience method for executing chain.

Return dictionary representation of agent.

Save the agent.

Utilize the LLM generate method for speed gains.