Spaces:
Runtime error
Runtime error
devopsmarc
commited on
Commit
·
3058a65
1
Parent(s):
cc56459
Initial Commit
Browse files- app.py +74 -0
- requirement.txt +7 -0
app.py
ADDED
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import os
|
2 |
+
os.environ["OPENAI_API_KEY"] = "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"
|
3 |
+
|
4 |
+
import openai
|
5 |
+
import os
|
6 |
+
import PyPDF2
|
7 |
+
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
|
8 |
+
from dotenv import load_dotenv, find_dotenv
|
9 |
+
|
10 |
+
_ = load_dotenv(find_dotenv())
|
11 |
+
|
12 |
+
openai.api_key = os.getenv('OPENAI_API_KEY')
|
13 |
+
|
14 |
+
def get_completion(prompt, model="gpt-3.5-turbo", temperature=0, max_tokens=500):
|
15 |
+
messages = [{"role": "user", "content": prompt}]
|
16 |
+
|
17 |
+
response = openai.ChatCompletion.create(
|
18 |
+
model=model,
|
19 |
+
temperature=temperature,
|
20 |
+
max_tokens=max_tokens,
|
21 |
+
messages=messages,
|
22 |
+
)
|
23 |
+
|
24 |
+
return response.choices[0].message["content"]
|
25 |
+
|
26 |
+
def generate_prompt(text, format="text"):
|
27 |
+
prompt = f"""Play as an AI HR recruiter specialist and extract all the jobs as a list with :
|
28 |
+
1. Job Title
|
29 |
+
2. Location
|
30 |
+
3. Educations as list
|
31 |
+
4. Experiences as list
|
32 |
+
5. Skills and Competences as list
|
33 |
+
6. Functions and Tasks as list
|
34 |
+
7. Return the result in {format} format \
|
35 |
+
here the text \
|
36 |
+
``` {text}```
|
37 |
+
|
38 |
+
|
39 |
+
"""
|
40 |
+
return prompt
|
41 |
+
|
42 |
+
import PyPDF2
|
43 |
+
|
44 |
+
def read_pdf(file_path):
|
45 |
+
pdf_file = open(file_path, 'rb')
|
46 |
+
pdf_reader = PyPDF2.PdfReader(pdf_file)
|
47 |
+
text = ""
|
48 |
+
for page in pdf_reader.pages:
|
49 |
+
text += page.extract_text()
|
50 |
+
pdf_file.close()
|
51 |
+
return text
|
52 |
+
|
53 |
+
import PyPDF2
|
54 |
+
def summarize(input):
|
55 |
+
text = read_pdf(input.name)
|
56 |
+
prompt = generate_prompt(text)
|
57 |
+
response = get_completion(prompt)
|
58 |
+
return response
|
59 |
+
|
60 |
+
import gradio as gr
|
61 |
+
gr.close_all()
|
62 |
+
demo = gr.Interface(fn=summarize, inputs="file", outputs="text",title="Experiences And Skills Extraction From PDF file with gpt-3.5-turbo")
|
63 |
+
demo.launch()
|
64 |
+
|
65 |
+
#++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
|
66 |
+
#++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
|
67 |
+
|
68 |
+
import gradio as gr
|
69 |
+
|
70 |
+
def colorize_text(text):
|
71 |
+
return "<span style='color:red;background-color:orange'>" + text + "</span>" "<span style='color:white;background-color:red'>" + "Experiences" + "</span>"
|
72 |
+
|
73 |
+
iface = gr.Interface(fn=colorize_text, inputs="text", outputs="html")
|
74 |
+
iface.launch()
|
requirement.txt
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
openai==0.27.0
|
2 |
+
transformers==4.31.0
|
3 |
+
sentencepiece==0.1.99
|
4 |
+
PyPDF2==3.0.1
|
5 |
+
python-dotenv==1.0.0
|
6 |
+
gradio==3.39.0
|
7 |
+
torch==2.0.1
|