Text classifier is giving "internal error"

Questions
1

I am getting an internal error using the text classifier node.
I set it up and all nodes prior are functional. I setup my credentials for Gemini successfully. But, when I run the workflow or test the node itself, I get “internal error” error message. I have no idea what to do about that. Is this a product bug or is there a fix for this I can’t foresee? Thanks!

What is the error message (if any)?

Screenshot 2024-08-04 at 10.01.35 PM
Screenshot 2024-08-04 at 10.01.35 PM1274×1332 53.5 KB

Please share your workflow

Screenshot 2024-08-04 at 10.07.06 PM
Screenshot 2024-08-04 at 10.07.06 PM2588×776 162 KB

I’m overall trying to 1. download a pdf from an email filtered by labels. 2. extract text from the pdf and turn it into JSON 3. break down the text into fields and 4. eventually append the data to a google sheet (another node not working but that’s not the specific topic I’m posting this for).


## Information on your n8n setup
- **n8n version:**
- **Database (default: SQLite):**
- **n8n EXECUTIONS_PROCESS setting (default: own, main):**
- **Running n8n via (Docker, npm, n8n cloud, desktop app):**
- **Operating system:**
2

It looks like your topic is missing some important information. Could you provide the following if applicable.

  • n8n version:
  • Database (default: SQLite):
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app):
  • Operating system:
3

Hey @Erin_cropcircle. Welcome :wave:

The Text Classifier node is intended for categorising incoming data and routing the workflow accordingly so it isn’t the best fit for what you want to do with the PDF data.

If you’re on v1.53.0+, check out the Information Extractor node instead which is designed for your use-case.

Otherwise if you’re on <v1.53.0, you can create your own information extractor by attaching a “structured output parser” to your LLM node like so:

Hope this helps!