ehdwns1516 commited on
Commit
c8ec3a1
1 Parent(s): f12c692

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +48 -0
README.md ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # ehdwns1516/gpt3-kor-based_gpt2_review_SR5
2
+
3
+ * This model has been trained Korean dataset as a star of 5 in the [naver shopping reivew dataset](https://github.com/bab2min/corpus/tree/master/sentiment).
4
+ * Input text what you want to generate review.
5
+ * If the context is longer than 1200 characters, the context may be cut in the middle and the result may not come out well.
6
+
7
+ review generator DEMO: [Ainize DEMO](https://main-review-generator-ehdwns1516.endpoint.ainize.ai/)
8
+
9
+ review generator API: [Ainize API](https://ainize.web.app/redirect?git_repo=https://github.com/ehdwns1516/review_generator)
10
+
11
+ ## Model links for each 1 to 5 star
12
+ * [ehdwns1516/gpt3-kor-based_gpt2_review_SR1](https://huggingface.co/ehdwns1516/gpt3-kor-based_gpt2_review_SR1)
13
+ * [ehdwns1516/gpt3-kor-based_gpt2_review_SR2](https://huggingface.co/ehdwns1516/gpt3-kor-based_gpt2_review_SR2)
14
+ * [ehdwns1516/gpt3-kor-based_gpt2_review_SR3](https://huggingface.co/ehdwns1516/gpt3-kor-based_gpt2_review_SR3)
15
+ * [ehdwns1516/gpt3-kor-based_gpt2_review_SR4](https://huggingface.co/ehdwns1516/gpt3-kor-based_gpt2_review_SR4)
16
+ * [ehdwns1516/gpt3-kor-based_gpt2_review_SR5](https://huggingface.co/ehdwns1516/gpt3-kor-based_gpt2_review_SR5)
17
+
18
+ ## Overview
19
+
20
+ Language model: [gpt3-kor-small_based_on_gpt2](https://huggingface.co/kykim/gpt3-kor-small_based_on_gpt2)
21
+
22
+ Language: Korean
23
+
24
+ Training data: review_body dataset with a star of 5 in the [naver shopping reivew dataset](https://github.com/bab2min/corpus/tree/master/sentiment).
25
+
26
+ Code: See [Ainize Workspace](https://ainize.ai/workspace/create?imageId=hnj95592adzr02xPTqss&git=https://github.com/ehdwns1516/gpt2_review_fine-tunning_note)
27
+
28
+ ## Usage
29
+ ## In Transformers
30
+
31
+ ```
32
+ from transformers import AutoTokenizer, AutoModelWithLMHead
33
+
34
+ tokenizer = AutoTokenizer.from_pretrained("ehdwns1516/gpt3-kor-based_gpt2_review_SR5")
35
+
36
+ model = AutoModelWithLMHead.from_pretrained("ehdwns1516/gpt3-kor-based_gpt2_review_SR5")
37
+
38
+ generator = pipeline(
39
+ "text-generation",
40
+ model="ehdwns1516/gpt3-kor-based_gpt2_review_SR5",
41
+ tokenizer=tokenizer
42
+ )
43
+
44
+ context = "your context"
45
+
46
+ result = dict()
47
+ result[0] = generator(context)[0]
48
+ ```