Spaces:
Runtime error
Runtime error
cahya
commited on
Commit
•
501a925
0
Parent(s):
Duplicate from cahya/indochat
Browse files- .gitattributes +34 -0
- README.md +14 -0
- app.py +92 -0
- requirements.txt +3 -0
.gitattributes
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
28 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
29 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
30 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
31 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
32 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
33 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
34 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
title: Indochat
|
3 |
+
emoji: 🇮🇩
|
4 |
+
colorFrom: purple
|
5 |
+
colorTo: red
|
6 |
+
sdk: gradio
|
7 |
+
sdk_version: 3.18.0
|
8 |
+
app_file: app.py
|
9 |
+
pinned: false
|
10 |
+
license: creativeml-openrail-m
|
11 |
+
duplicated_from: cahya/indochat
|
12 |
+
---
|
13 |
+
|
14 |
+
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
app.py
ADDED
@@ -0,0 +1,92 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
import os
|
3 |
+
from mtranslate import translate
|
4 |
+
import requests
|
5 |
+
|
6 |
+
HF_AUTH_TOKEN = os.environ.get("HF_AUTH_TOKEN")
|
7 |
+
indochat_api = 'https://cahya-indonesian-whisperer.hf.space/api/indochat/v1'
|
8 |
+
indochat_api_auth_token = os.getenv("INDOCHAT_API_AUTH_TOKEN", "")
|
9 |
+
|
10 |
+
def get_answer(user_input, decoding_method, num_beams, top_k, top_p, temperature, repetition_penalty, penalty_alpha):
|
11 |
+
print(user_input, decoding_method, top_k, top_p, temperature, repetition_penalty, penalty_alpha)
|
12 |
+
headers = {'Authorization': 'Bearer ' + indochat_api_auth_token}
|
13 |
+
data = {
|
14 |
+
"text": user_input,
|
15 |
+
"min_length": len(user_input) + 50,
|
16 |
+
"max_length": 300,
|
17 |
+
"decoding_method": decoding_method,
|
18 |
+
"num_beams": num_beams,
|
19 |
+
"top_k": top_k,
|
20 |
+
"top_p": top_p,
|
21 |
+
"temperature": temperature,
|
22 |
+
"seed": -1,
|
23 |
+
"repetition_penalty": repetition_penalty,
|
24 |
+
"penalty_alpha": penalty_alpha
|
25 |
+
}
|
26 |
+
r = requests.post(indochat_api, headers=headers, data=data)
|
27 |
+
if r.status_code == 200:
|
28 |
+
result = r.json()
|
29 |
+
answer = result["generated_text"]
|
30 |
+
user_input_en = translate(user_input, "en", "id")
|
31 |
+
answer_en = translate(answer, "en", "id")
|
32 |
+
return [(f"{user_input}\n", None), (answer, "")], \
|
33 |
+
[(f"{user_input_en}\n", None), (answer_en, "")]
|
34 |
+
else:
|
35 |
+
return "Error: " + r.text
|
36 |
+
|
37 |
+
|
38 |
+
css = """
|
39 |
+
#answer_id span {white-space: pre-line}
|
40 |
+
#answer_id span.label {display: none}
|
41 |
+
#answer_en span {white-space: pre-line}
|
42 |
+
#answer_en span.label {display: none}
|
43 |
+
"""
|
44 |
+
|
45 |
+
with gr.Blocks(css=css) as demo:
|
46 |
+
with gr.Row():
|
47 |
+
gr.Markdown("""## IndoChat
|
48 |
+
|
49 |
+
A Prove of Concept of a multilingual Chatbot (in this case a bilingual, English and Indonesian), fine-tuned with
|
50 |
+
multilingual instructions dataset. The base model is a GPT2-Medium (340M params) which was pretrained with 75GB
|
51 |
+
of Indonesian and English dataset, where English part is only less than 1% of the whole dataset.
|
52 |
+
""")
|
53 |
+
with gr.Row():
|
54 |
+
with gr.Column():
|
55 |
+
user_input = gr.inputs.Textbox(placeholder="",
|
56 |
+
label="Ask me something in Indonesian or English",
|
57 |
+
default="Bagaimana cara mendidik anak supaya tidak berbohong?")
|
58 |
+
decoding_method = gr.inputs.Dropdown(["Beam Search", "Sampling", "Contrastive Search"],
|
59 |
+
default="Sampling", label="Decoding Method")
|
60 |
+
num_beams = gr.inputs.Slider(label="Number of beams for beam search",
|
61 |
+
default=1, minimum=1, maximum=10, step=1)
|
62 |
+
top_k = gr.inputs.Slider(label="Top K",
|
63 |
+
default=30, maximum=50, minimum=1, step=1)
|
64 |
+
top_p = gr.inputs.Slider(label="Top P", default=0.9, step=0.05, minimum=0.1, maximum=1.0)
|
65 |
+
temperature = gr.inputs.Slider(label="Temperature", default=0.5, step=0.05, minimum=0.1, maximum=1.0)
|
66 |
+
repetition_penalty = gr.inputs.Slider(label="Repetition Penalty", default=1.1, step=0.05, minimum=1.0, maximum=2.0)
|
67 |
+
penalty_alpha = gr.inputs.Slider(label="The penalty alpha for contrastive search",
|
68 |
+
default=0.5, step=0.05, minimum=0.05, maximum=1.0)
|
69 |
+
with gr.Row():
|
70 |
+
button_generate_story = gr.Button("Submit")
|
71 |
+
with gr.Column():
|
72 |
+
# generated_answer = gr.Textbox()
|
73 |
+
generated_answer = gr.HighlightedText(
|
74 |
+
elem_id="answer_id",
|
75 |
+
label="Generated Text",
|
76 |
+
combine_adjacent=True,
|
77 |
+
css="#htext span {white-space: pre-line}",
|
78 |
+
).style(color_map={"": "blue", "-": "green"})
|
79 |
+
generated_answer_en = gr.HighlightedText(
|
80 |
+
elem_id="answer_en",
|
81 |
+
label="Translation",
|
82 |
+
combine_adjacent=True,
|
83 |
+
).style(color_map={"": "blue", "-": "green"})
|
84 |
+
with gr.Row():
|
85 |
+
gr.Markdown("![visitor badge](https://visitor-badge.glitch.me/badge?page_id=cahya_indochat)")
|
86 |
+
|
87 |
+
button_generate_story.click(get_answer,
|
88 |
+
inputs=[user_input, decoding_method, num_beams, top_k, top_p, temperature,
|
89 |
+
repetition_penalty, penalty_alpha],
|
90 |
+
outputs=[generated_answer, generated_answer_en])
|
91 |
+
|
92 |
+
demo.launch(enable_queue=False)
|
requirements.txt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
torch
|
2 |
+
transformers
|
3 |
+
mtranslate
|