Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,120 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
+
datasets:
|
4 |
+
- hakurei/open-instruct-v1
|
5 |
+
tags:
|
6 |
+
- generated_from_trainer
|
7 |
+
- alpaca
|
8 |
+
- self-instruct
|
9 |
+
- instruction generation
|
10 |
+
- instructiongen
|
11 |
+
|
12 |
+
widget:
|
13 |
+
- text: >-
|
14 |
+
You'll need to start by choosing the right venue. Consider the type of
|
15 |
+
atmosphere and the size of the area that will be suitable for the number
|
16 |
+
of guests you plan to invite. Choose the right decorations based on your
|
17 |
+
brother's interests, such as balloons in his favorite colors, banners, and
|
18 |
+
streamers. Next, decide on the food and drinks, making sure they are tasty
|
19 |
+
and appropriate for the occasion. Then decide on the other games, music,
|
20 |
+
and entertainment that will make the party memorable. Finally, involve
|
21 |
+
your brother's friends and family to help create the perfect surprise.
|
22 |
+
example_title: birthday party
|
23 |
+
- text: 1) cookies and cream 2) chocolate chip 3) mint chip 4) oreo
|
24 |
+
example_title: ice cream
|
25 |
+
- text: >-
|
26 |
+
Start by selecting a scale model of a building that fits the theme. Use a
|
27 |
+
hobby knife and glue to cut and assemble the model into a ruined or
|
28 |
+
abandoned version of itself, adding details like broken windows and
|
29 |
+
graffiti. Create a base for the diorama using foam, plaster, or other
|
30 |
+
materials, and paint it to resemble a ruined street or sidewalk. Add
|
31 |
+
miniature vehicles, debris, and figures to complete the scene, and use
|
32 |
+
weathering techniques like dry brushing and rust washes to add realism.
|
33 |
+
Display the diorama in a shadow box or other protective case to showcase
|
34 |
+
your work.
|
35 |
+
example_title: Miniature diorama creation
|
36 |
+
- text: >-
|
37 |
+
Start by selecting clothing that is futuristic and edgy, such as leather
|
38 |
+
jackets, neon-colored accessories, and tech-inspired patterns. Add
|
39 |
+
accessories like goggles, cybernetic implants, and LED lights to enhance
|
40 |
+
the cyberpunk vibe. Use makeup and body paint to create a futuristic look,
|
41 |
+
such as metallic skin or neon makeup. Consider adding functional elements
|
42 |
+
to your costume, such as a built-in backpack or hidden pockets for your
|
43 |
+
tech gadgets. Finally, practice your confident walk and embrace your inner
|
44 |
+
cyberpunk for a memorable and immersive costume experience.
|
45 |
+
example_title: Cyberpunk costume design
|
46 |
+
- text: >-
|
47 |
+
Start by creating a base terrain with mountains, valleys, and other
|
48 |
+
natural features. Use fractal noise and displacement mapping to add
|
49 |
+
texture and detail to the terrain, and experiment with different materials
|
50 |
+
like rock, grass, and water. Add surreal elements like floating islands,
|
51 |
+
giant mushrooms, or impossible geometry to create a dreamlike atmosphere.
|
52 |
+
Use lighting and color grading to enhance the mood and tone of the scene,
|
53 |
+
and render the final image at a high resolution for maximum impact. Share
|
54 |
+
your surreal landscape with the world and inspire others to explore the
|
55 |
+
possibilities of 3D art.
|
56 |
+
example_title: Surreal 3D landscape creation
|
57 |
+
- text: >-
|
58 |
+
Start by setting a realistic goal and creating a training plan. Build up
|
59 |
+
your mileage gradually over time, and incorporate cross-training and
|
60 |
+
strength exercises to prevent injury and improve endurance. Be sure to
|
61 |
+
stay hydrated and properly fuel your body with nutritious foods. Listen to
|
62 |
+
your body and adjust your training as needed to avoid overexertion or
|
63 |
+
burnout. Finally, taper your training in the weeks leading up to the race
|
64 |
+
to give your body time to rest and recover before the big day.
|
65 |
+
example_title: Marathon training
|
66 |
+
inference:
|
67 |
+
parameters:
|
68 |
+
max_length: 96
|
69 |
+
num_beams: 4
|
70 |
+
encoder_no_repeat_ngram_size: 4
|
71 |
---
|
72 |
+
|
73 |
+
|
74 |
+
# pszemraj/bart-base-open-instructiongen-v1
|
75 |
+
|
76 |
+
|
77 |
+
Instead of generating questions from text, generate instructions for LLMs!
|
78 |
+
|
79 |
+
Check out a basic demo on Spaces (https://huggingface.co/spaces/pszemraj/generate-instructions). You can find other models fine-tuned for instruction generation by [searching for the instructiongen tag](https://huggingface.co/models?other=instructiongen).
|
80 |
+
|
81 |
+
## Model description
|
82 |
+
|
83 |
+
This model is a fine-tuned version of [facebook/bart-base](https://huggingface.co/facebook/bart-base) on the hakurei/open-instruct-v1 dataset.
|
84 |
+
|
85 |
+
- This model **only** generates the `instruction` for arbitrary text (it **does not** provide `inputs` as well - look for models with `w-inputs` in the name).
|
86 |
+
- There was no validation split at the time of training, so no statistics here.
|
87 |
+
- Comparing the performance of this model with `pszemraj/bart-base-instructiongen` might give some indication of whether and how much dataset scaling is needed to produce "robust" instruction generators.
|
88 |
+
- If you notice any trends, feel free to reach out! would be happy to hear about it.
|
89 |
+
|
90 |
+
## Training and evaluation data
|
91 |
+
|
92 |
+
See `hakurei/open-instruct-v1`. This model was trained on the dataset "backwards", i.e. the model was given the `output` column as input and trained to predict `instruction`.
|
93 |
+
|
94 |
+
## Training procedure
|
95 |
+
|
96 |
+
### Training hyperparameters
|
97 |
+
|
98 |
+
The following hyperparameters were used during training:
|
99 |
+
- learning_rate: 8e-05
|
100 |
+
- train_batch_size: 16
|
101 |
+
- eval_batch_size: 8
|
102 |
+
- seed: 42
|
103 |
+
- distributed_type: multi-GPU
|
104 |
+
- gradient_accumulation_steps: 2
|
105 |
+
- total_train_batch_size: 32
|
106 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
107 |
+
- lr_scheduler_type: cosine
|
108 |
+
- lr_scheduler_warmup_ratio: 0.03
|
109 |
+
- num_epochs: 2.0
|
110 |
+
|
111 |
+
### Training results
|
112 |
+
|
113 |
+
|
114 |
+
|
115 |
+
### Framework versions
|
116 |
+
|
117 |
+
- Transformers 4.28.0.dev0
|
118 |
+
- Pytorch 2.0.0+cu118
|
119 |
+
- Datasets 2.9.0
|
120 |
+
- Tokenizers 0.12.1
|