-
Hi @PhilipMay,
thanks for releasing this German ELMo model :+1:
I've one question: would it be possible that you provide the `options.json` and `weights.hdf5` as separate downloads 🤔
I would…
-
From other discussions, I understand the token embeddings aren't trainable here, but a similar effect is achieved via a fully-connected layer on top of the embeddings.
I am using both glove embeddi…
-
To be removed, once it is done: Please add the appropriate label to this ticket, e.g. feature or enhancement.
**Is your feature/enhancement request related to a problem? Please describe.**
A clear…
-
If `embeddings_storage_mode = 'cpu'` is selected during training, all word embedding tensors are moved to CPU memory and are at each epoch copied over to GPU memory when processing a mini-batch. It's …
-
It appears that argument `max_tokens_per_doc` has no effect when creating a `ClassificationCorpus` if `in_memory=False` (which is the default for `ClassificationCorpus`).
Looking at `Classification…
-
As the corpus of #12 grows, and we start using the full article text, instead of just its title, the training gets really slow, and a laptop is not the ideal way. How much faster is the training with …
dcabo updated
5 years ago
-
Following the tutorial on how to train the embeddings in my own corpus I got an issue:
my structure is exactly the same as in the example:
├── corpus
│  ├── test.txt
│  ├── train
│  │  ├── …
-
Hi,
I am following [this tutorial](https://github.com/zalandoresearch/flair/blob/master/resources/docs/TUTORIAL_9_TRAINING_LM_EMBEDDINGS.md) to train Flair embeddings on my corpus. My folder structur…
-
In your code there are multiple references to ELMo, indicating that you experimented with contextual embeddings.
Can you share any of your results using ELMo embeddings?
I am currently getting f1…
-
**Describe the bug**
I was using v0.4.2 of flair and had some tests that I used while developing on a AWS instance with 4GB RAM. Before, running the tests on training did not consume all the RAM. I'v…