-
Hi!
I'm experimenting with training a TextClassifier on the 20 newsgroups dataset. My interest is the performance of the pure flair embeddings, especially.
With my example code, though, neither …
-
Hi FLAIR Team,
I have been evaluating FLAIR for a NER task and got nice results (thanks!). I used the default train-dev-test setup for my experiments.
However, I want to use the NER model in a downs…
-
I would like to confirm whether the pretrained weights available in the README are just from "masked input node" prediction, and *not* of the final trained MAT. I assume this is the case because it sk…
-
**Describe the bug**
I ran into the 'cpu+0' issue described [here](https://github.com/flairNLP/flair/issues/1388) and i did pip install --upgrade git+https://github.com/flairNLP/flair.git so i am on …
-
Hi,
a few weeks before the BERT paper (which seems to got all the attention), the "Semi-Supervised Sequence Modeling with Cross-View Training" [1],[2] paper was published. It shows a nice semi-supe…
-
Hello guys!
I followed the instructions of the tutorials to construct my own training data for NER.
All I want to do is train a custom NER model with my own data.
My data looks like this:
> Ge…
-
Generating paragraph embeddings based weighted sentence embeddings, sentence weights are `NSP+CS` score wrt ChannelMind vectors.
![image](https://user-images.githubusercontent.com/16121695/5783193…
-
We need the functionality to calculate `b = index_add(a, indices, value)`, which mimics the outcome of `a[indices] += value`.
This is similar to the `tensor_scatter_nd_add` in TF: https://www.tenso…
-
Voici l'article que je vous ai brièvement présenté de Lê Nguyen Hoang, écrit en 2017 et revu en 2019, pour mieux coller à ses réflexions et à la publication de son livre sur la question [Le fabuleux c…
-
Can we modify the Thursday and Friday tabs to be chronological ordering?
Ideally, this would be:
Session Time (EST), Content Type (Tutorial, Keynote, Doctoral Symposium, Research Talk, Workshop Ta…