Janis Meyer

icon company self employed icon location Hamburg, Germany icon location Hi, I am Janis, I am particularly interested in the question of how machine learning can be used to extract information from complex structured documents.

Results 5 issues of Janis Meyer

Implementing Accuracy, F1 for evaluating sequence and token classifier

**Enhancement** 🚀 Implementing accuracy, f1 score and confusion matrix for sequence and token classifier **Motivation** 💪 Reducing dependencies, e.g. sklearn **Alternatives** ⚖️ Accuracy, confusion currently use sklearn **Additional context** 🧬

design

Upgrade `catalogue` to latest release

**Bug** 💥 `catalogue==2.0.7` is not compatible with `importlib.metadata>=5.0.0`. Use this PR to upgrade various other third party library dependencies **Expected behavior** 🧮 Consistent dependency of both libraries. **Screenshots** 🖼 If...

bug

Reducing dependency complexity: replace `jsonlines` and `pyyaml` and built-in `json` with explosion/`srsly`

**Enhancement** 🚀 Replacing `jsonlines` and `pyyaml` with `srsly` **Motivation** 💪 `srsly` is a serialization package that features loading of the most important formats in **deep**doctection: `json`, `jsonlines` and `yaml`. Replacing...

enhancement

Adding a pipeline component for zero shot LLM document classification and NER

**Enhancement** 🚀 With support of LLM integration libraries like Llama index or Langchain it is fairly easy to add a pipeline component for zero shot classification or entity recognition using...

enhancement

`MisformedRegexError` when using the `gen` with a regular expression

4 comment

**The bug** I encounter an `MisformedRegexError` when using the `gen` with a regular expression. The error tells me `The regex is misformed here. Regex: |` even though the regex I...