Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,114 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
datasets:
|
3 |
+
- EleutherAI/pile
|
4 |
+
language:
|
5 |
+
- en
|
6 |
+
pipeline_tag: fill-mask
|
7 |
+
tags:
|
8 |
+
- summarization
|
9 |
+
- translation
|
10 |
+
---
|
11 |
+
|
12 |
+
# Model Card for T5v2 Base
|
13 |
+
|
14 |
+
# Table of Contents
|
15 |
+
|
16 |
+
1. [Model Details](#model-details)
|
17 |
+
2. [Uses](#uses)
|
18 |
+
3. [Bias, Risks, and Limitations](#bias-risks-and-limitations)
|
19 |
+
4. [Training Details](#training-details)
|
20 |
+
5. [Evaluation](#evaluation)
|
21 |
+
6. [Environmental Impact](#environmental-impact)
|
22 |
+
7. [Citation](#citation)
|
23 |
+
8. [Model Card Authors](#model-card-authors)
|
24 |
+
9. [How To Get Started With the Model](#how-to-get-started-with-the-model)
|
25 |
+
|
26 |
+
# Model Details
|
27 |
+
|
28 |
+
## Model Description
|
29 |
+
|
30 |
+
More information needed.
|
31 |
+
# Uses
|
32 |
+
|
33 |
+
## Direct Use and Downstream Use
|
34 |
+
|
35 |
+
More information needed.
|
36 |
+
|
37 |
+
## Out-of-Scope Use
|
38 |
+
|
39 |
+
More information needed.
|
40 |
+
|
41 |
+
# Bias, Risks, and Limitations
|
42 |
+
|
43 |
+
More information needed.
|
44 |
+
|
45 |
+
## Recommendations
|
46 |
+
|
47 |
+
More information needed.
|
48 |
+
|
49 |
+
# Training Details
|
50 |
+
|
51 |
+
## Training Data
|
52 |
+
|
53 |
+
The model was pre-trained on the Pile using an unsupervised denoising objective,
|
54 |
+
## Training Procedure
|
55 |
+
|
56 |
+
More information needed.
|
57 |
+
|
58 |
+
# Evaluation
|
59 |
+
|
60 |
+
## Testing Data, Factors & Metrics
|
61 |
+
|
62 |
+
More information needed.
|
63 |
+
## Results
|
64 |
+
|
65 |
+
More information needed.
|
66 |
+
|
67 |
+
# Environmental Impact
|
68 |
+
|
69 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
70 |
+
|
71 |
+
- **Hardware Type:** Google Cloud TPU Pods
|
72 |
+
- **Hours used:** More information needed
|
73 |
+
- **Cloud Provider:** GCP
|
74 |
+
- **Compute Region:** More information needed
|
75 |
+
- **Carbon Emitted:** More information needed
|
76 |
+
|
77 |
+
# Citation
|
78 |
+
|
79 |
+
**BibTeX:**
|
80 |
+
|
81 |
+
```bibtex
|
82 |
+
@article{2024t5v2,
|
83 |
+
author = {Lintang Sutawika and Aran Komatsuzaki and Colin Raffel},
|
84 |
+
title = {T5v2, an update of T5},
|
85 |
+
year = {2024},
|
86 |
+
url = {}
|
87 |
+
}
|
88 |
+
```
|
89 |
+
|
90 |
+
# How to Get Started with the Model
|
91 |
+
|
92 |
+
Use the code below to get started with the model.
|
93 |
+
|
94 |
+
<details>
|
95 |
+
<summary> Click to expand </summary>
|
96 |
+
|
97 |
+
```python
|
98 |
+
from transformers import UMT5Tokenizer, UMT5Model
|
99 |
+
|
100 |
+
tokenizer = UMT5Tokenizer.from_pretrained("EleutherAI/t5-v2-base")
|
101 |
+
model = UMT5Model.from_pretrained("EleutherAI/t5-v2-base")
|
102 |
+
|
103 |
+
input_ids = tokenizer(
|
104 |
+
"Studies have been shown that owning a dog is good for you", return_tensors="pt"
|
105 |
+
).input_ids # Batch size 1
|
106 |
+
decoder_input_ids = tokenizer("Studies show that", return_tensors="pt").input_ids # Batch size 1
|
107 |
+
|
108 |
+
# forward pass
|
109 |
+
outputs = model(input_ids=input_ids, decoder_input_ids=decoder_input_ids)
|
110 |
+
last_hidden_states = outputs.last_hidden_state
|
111 |
+
```
|
112 |
+
|
113 |
+
|
114 |
+
</details>
|