Fix model card YAML frontmatter (valid pipeline_tag, clean metadata)
Browse files
README.md
CHANGED
|
@@ -1,94 +1,95 @@
|
|
| 1 |
-
---
|
| 2 |
-
pipeline_tag: feature-extraction
|
| 3 |
-
license: apache-2.0
|
| 4 |
-
base_model: facebook/bart-base
|
| 5 |
-
library_name: zeromodels
|
| 6 |
-
tags:
|
| 7 |
-
- keras
|
| 8 |
-
- zeromodels
|
| 9 |
-
- bart
|
| 10 |
-
-
|
| 11 |
-
-
|
| 12 |
-
-
|
| 13 |
-
-
|
| 14 |
-
-
|
| 15 |
-
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
|
|
| 60 |
-
|
|
| 61 |
-
| `
|
| 62 |
-
| `
|
| 63 |
-
| `
|
| 64 |
-
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
|
|
| 71 |
-
|
|
| 72 |
-
| `
|
| 73 |
-
| `
|
| 74 |
-
| `
|
| 75 |
-
|
| 76 |
-
|
| 77 |
-
|
| 78 |
-
|
| 79 |
-
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
-
|
| 86 |
-
-
|
| 87 |
-
-
|
| 88 |
-
-
|
| 89 |
-
|
| 90 |
-
|
| 91 |
-
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pipeline_tag: feature-extraction
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: facebook/bart-base
|
| 5 |
+
library_name: zeromodels
|
| 6 |
+
tags:
|
| 7 |
+
- keras
|
| 8 |
+
- zeromodels
|
| 9 |
+
- bart
|
| 10 |
+
- feature-extraction
|
| 11 |
+
- seq2seq
|
| 12 |
+
- arxiv:1910.13461
|
| 13 |
+
- pytorch
|
| 14 |
+
- jax
|
| 15 |
+
- tf
|
| 16 |
+
---
|
| 17 |
+
|
| 18 |
+
# Run BART with Keras 3: JAX, PyTorch, or TensorFlow
|
| 19 |
+
|
| 20 |
+
[](https://github.com/ZeroAIx/ZeroModels) [](https://zeroaix.github.io/ZeroModels/bart/)
|
| 21 |
+
|
| 22 |
+
# zeromodels/bart_base
|
| 23 |
+
|
| 24 |
+
Paper: [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension (arXiv:1910.13461)](https://arxiv.org/abs/1910.13461) · [HF Papers](https://huggingface.co/papers/1910.13461)
|
| 25 |
+
|
| 26 |
+
BART is a denoising seq2seq transformer: a bidirectional encoder (like BERT) and an autoregressive decoder (like GPT) trained to reconstruct corrupted text. It excels at summarization, translation, and other text-to-text tasks. Byte-level BPE tokenizer (shared with RoBERTa); the decoder starts from `</s>`.
|
| 27 |
+
|
| 28 |
+
For more details on the model, please go to the upstream [model card](https://huggingface.co/facebook/bart-base).
|
| 29 |
+
|
| 30 |
+
Pure-**Keras 3** conversion of [`facebook/bart-base`](https://huggingface.co/facebook/bart-base) for [zeromodels](https://github.com/ZeroAIx/ZeroModels). One implementation runs unmodified on **TensorFlow / Torch / JAX**.
|
| 31 |
+
|
| 32 |
+
This is a **conditional generation (base seq2seq)** checkpoint (`BartConditionalGenerate`). Other task heads load the shared backbone from this repo (start randomly initialized, ready for fine-tuning); fine-tuned task checkpoints load via the `hf:` prefix.
|
| 33 |
+
|
| 34 |
+
> Base checkpoint (not task fine-tuned): use it as a backbone (`BartModel`) for features, or fine-tune a task head.
|
| 35 |
+
|
| 36 |
+
## ✨ Quick start
|
| 37 |
+
|
| 38 |
+
```python
|
| 39 |
+
import os
|
| 40 |
+
os.environ["KERAS_BACKEND"] = "torch" # or "jax" / "tensorflow"
|
| 41 |
+
|
| 42 |
+
from zeromodels.models.bart import BartConditionalGenerate, BartTokenizer
|
| 43 |
+
|
| 44 |
+
model = BartConditionalGenerate.from_weights("zeromodels/bart_base")
|
| 45 |
+
tokenizer = BartTokenizer.from_weights("zeromodels/bart_base")
|
| 46 |
+
|
| 47 |
+
inputs = tokenizer('The quick brown fox jumps over the lazy dog.')
|
| 48 |
+
ids = model.generate(
|
| 49 |
+
inputs,
|
| 50 |
+
[[model.decoder_start_token_id]],
|
| 51 |
+
max_new_tokens=64,
|
| 52 |
+
eos_token_id=tokenizer.eos_token_id,
|
| 53 |
+
)
|
| 54 |
+
print(tokenizer.decode(ids[0], skip_special_tokens=True))
|
| 55 |
+
```
|
| 56 |
+
|
| 57 |
+
Load any BART variant the same way with `from_weights("zeromodels/<variant>")`:
|
| 58 |
+
|
| 59 |
+
| Variant | Hub | Task |
|
| 60 |
+
|---|---|---|
|
| 61 |
+
| `bart_base` | [`zeromodels/bart_base`](https://huggingface.co/zeromodels/bart_base) | conditional generation (base seq2seq) |
|
| 62 |
+
| `bart_large` | [`zeromodels/bart_large`](https://huggingface.co/zeromodels/bart_large) | conditional generation (base seq2seq) |
|
| 63 |
+
| `bart_large_cnn` | [`zeromodels/bart_large_cnn`](https://huggingface.co/zeromodels/bart_large_cnn) | summarization (CNN / DailyMail) |
|
| 64 |
+
| `bart_large_xsum` | [`zeromodels/bart_large_xsum`](https://huggingface.co/zeromodels/bart_large_xsum) | extreme summarization (XSum, one-sentence) |
|
| 65 |
+
|
| 66 |
+
## Available classes
|
| 67 |
+
|
| 68 |
+
Load any of these from this repo with `from_weights("zeromodels/bart_base")` (or on the fly via the `hf:` prefix). The pretrained backbone is shared; task heads not stored in this checkpoint start randomly initialized, ready for fine-tuning (or load a `hf:` fine-tune).
|
| 69 |
+
|
| 70 |
+
| Class | Task |
|
| 71 |
+
|---|---|
|
| 72 |
+
| `BartModel` | Encoder-decoder backbone |
|
| 73 |
+
| `BartConditionalGenerate` | Conditional generation (summarization / seq2seq) |
|
| 74 |
+
| `BartSequenceClassify` | Sequence classification (e.g. NLI / zero-shot) |
|
| 75 |
+
| `BartQnA` | Extractive question answering |
|
| 76 |
+
|
| 77 |
+
```python
|
| 78 |
+
from zeromodels.models.bart import BartSequenceClassify
|
| 79 |
+
# zero-shot / NLI fine-tune loads on the fly via the hf: prefix
|
| 80 |
+
model = BartSequenceClassify.from_weights("hf:facebook/bart-large-mnli")
|
| 81 |
+
```
|
| 82 |
+
|
| 83 |
+
## Tips
|
| 84 |
+
|
| 85 |
+
- Set `KERAS_BACKEND` **before** importing Keras / zeromodels.
|
| 86 |
+
- Prefer `BartTokenizer.from_weights(...)` so the byte-level BPE matches.
|
| 87 |
+
- BART's decoder starts from `</s>` (`decoder_start_token_id = 2`); pass `eos_token_id=tokenizer.eos_token_id` to stop generation.
|
| 88 |
+
- See the [BART docs](https://zeroaix.github.io/ZeroModels/bart/) and [Loading Weights](https://zeroaix.github.io/ZeroModels/loading_weights/).
|
| 89 |
+
- Community / upstream safetensors still work via the `hf:` prefix, e.g. `BartConditionalGenerate.from_weights("hf:facebook/bart-base")`.
|
| 90 |
+
|
| 91 |
+
## Special Thanks
|
| 92 |
+
|
| 93 |
+
A huge thank you to the Meta AI (FAIR) authors for creating and releasing BART.
|
| 94 |
+
|
| 95 |
+
License: apache-2.0.
|