Felix Reichel

> @R4ZZ3 > Does the traditional translate() method work for you? > > The test_multi_process_translation.py script makes only sense if you want to translate using multiple GPUs. Could you explain,...

I ran into the very same issue as there is a json code block generated in the output examples in the prompts which cannot be read in properly with `json.loads`...

@Jeerhz i kept the examples but manually deleted the json code block instructions that were generated while adapting, something like `"classification": "```json {\n \"TP\": ..."`

@Jeerhz that's why: https://github.com/explodinggradients/ragas/blob/95d8318abdc9c38682a9c1da6e6c7781e69fade1/src/ragas/llms/prompt.py#L122 the prompts are properly converted to dict including the ticks and then stored as json file but when reading them with the method mentioned above it...

@jjmachan for me it only happens when reading the prompts in again after they were stored as json file. Hence, my issue isn't totally related to this one but returns...

@Flamefire Thanks for providing the pylint plugin for tensorflow. I tested it and recognized, that the `unable to import ...` errors were completely resolved. However, any module not found errors...