JingyaHuang
commited on
Commit
•
a4a9d1f
1
Parent(s):
cce4619
Upload gpt2 ONNX models
Browse files- README.md +53 -0
- config.json +39 -0
- decoder_model.onnx +3 -0
- decoder_with_past_model.onnx +3 -0
- tokenizer.json +0 -0
README.md
CHANGED
@@ -1,3 +1,56 @@
|
|
1 |
---
|
|
|
|
|
|
|
|
|
2 |
license: mit
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
language: en
|
3 |
+
tags:
|
4 |
+
- exbert
|
5 |
+
|
6 |
license: mit
|
7 |
---
|
8 |
+
|
9 |
+
|
10 |
+
# GPT-2
|
11 |
+
|
12 |
+
Test the whole generation capabilities here: https://transformer.huggingface.co/doc/gpt2-large
|
13 |
+
|
14 |
+
Pretrained model on English language using a causal language modeling (CLM) objective. It was introduced in
|
15 |
+
[this paper](https://d4mucfpksywv.cloudfront.net/better-language-models/language_models_are_unsupervised_multitask_learners.pdf)
|
16 |
+
and first released at [this page](https://openai.com/blog/better-language-models/).
|
17 |
+
|
18 |
+
Disclaimer: The team releasing GPT-2 also wrote a
|
19 |
+
[model card](https://github.com/openai/gpt-2/blob/master/model_card.md) for their model. Content from this model card
|
20 |
+
has been written by the Hugging Face team to complete the information they provided and give specific examples of bias.
|
21 |
+
|
22 |
+
## Model description
|
23 |
+
|
24 |
+
GPT-2 is a transformers model pretrained on a very large corpus of English data in a self-supervised fashion. This
|
25 |
+
means it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots
|
26 |
+
of publicly available data) with an automatic process to generate inputs and labels from those texts. More precisely,
|
27 |
+
it was trained to guess the next word in sentences.
|
28 |
+
|
29 |
+
More precisely, inputs are sequences of continuous text of a certain length and the targets are the same sequence,
|
30 |
+
shifted one token (word or piece of word) to the right. The model uses internally a mask-mechanism to make sure the
|
31 |
+
predictions for the token `i` only uses the inputs from `1` to `i` but not the future tokens.
|
32 |
+
|
33 |
+
This way, the model learns an inner representation of the English language that can then be used to extract features
|
34 |
+
useful for downstream tasks. The model is best at what it was pretrained for however, which is generating texts from a
|
35 |
+
prompt.
|
36 |
+
|
37 |
+
## Intended uses & limitations
|
38 |
+
|
39 |
+
You can use the raw model for text generation or fine-tune it to a downstream task. See the
|
40 |
+
[model hub](https://huggingface.co/models?filter=gpt2) to look for fine-tuned versions on a task that interests you.
|
41 |
+
|
42 |
+
### How to use
|
43 |
+
|
44 |
+
Here is how to use the ONNX models of gpt2 to get the features of a given text:
|
45 |
+
|
46 |
+
```python
|
47 |
+
from transformers import AutoTokenizer, pipeline
|
48 |
+
from optimum.onnxruntime import ORTModelForCausalLM
|
49 |
+
|
50 |
+
tokenizer = AutoTokenizer.from_pretrained("gpt2")
|
51 |
+
model = ORTModelForCausalLM.from_pretrained("gpt2", from_transformers=True)
|
52 |
+
onnx_gen = pipeline("text-generation", model=model, tokenizer=tokenizer)
|
53 |
+
|
54 |
+
text = "My name is Philipp and I live in Germany."
|
55 |
+
gen = onnx_gen(text)
|
56 |
+
```
|
config.json
ADDED
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "gpt2",
|
3 |
+
"activation_function": "gelu_new",
|
4 |
+
"architectures": [
|
5 |
+
"GPT2LMHeadModel"
|
6 |
+
],
|
7 |
+
"attn_pdrop": 0.1,
|
8 |
+
"bos_token_id": 50256,
|
9 |
+
"embd_pdrop": 0.1,
|
10 |
+
"eos_token_id": 50256,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"layer_norm_epsilon": 1e-05,
|
13 |
+
"model_type": "gpt2",
|
14 |
+
"n_ctx": 1024,
|
15 |
+
"n_embd": 768,
|
16 |
+
"n_head": 12,
|
17 |
+
"n_inner": null,
|
18 |
+
"n_layer": 12,
|
19 |
+
"n_positions": 1024,
|
20 |
+
"pad_token_id": 0,
|
21 |
+
"reorder_and_upcast_attn": false,
|
22 |
+
"resid_pdrop": 0.1,
|
23 |
+
"scale_attn_by_inverse_layer_idx": false,
|
24 |
+
"scale_attn_weights": true,
|
25 |
+
"summary_activation": null,
|
26 |
+
"summary_first_dropout": 0.1,
|
27 |
+
"summary_proj_to_labels": true,
|
28 |
+
"summary_type": "cls_index",
|
29 |
+
"summary_use_proj": true,
|
30 |
+
"task_specific_params": {
|
31 |
+
"text-generation": {
|
32 |
+
"do_sample": true,
|
33 |
+
"max_length": 50
|
34 |
+
}
|
35 |
+
},
|
36 |
+
"transformers_version": "4.24.0",
|
37 |
+
"use_cache": true,
|
38 |
+
"vocab_size": 50257
|
39 |
+
}
|
decoder_model.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bbb5a54bb827ae4bff3f0c4524a5d482fedbf914a77b669944f3d492b88b9e85
|
3 |
+
size 653447720
|
decoder_with_past_model.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:241d4ab52d60593730054d97ee3a20c30eee5a441239b6345e7ec78da4b1a9e8
|
3 |
+
size 653452603
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|