|
from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline |
|
model = AutoModelForSequenceClassification.from_pretrained("savasy/bert-base-turkish-sentiment-cased") |
|
tokenizer = AutoTokenizer.from_pretrained("savasy/bert-base-turkish-sentiment-cased") |
|
sa= pipeline("sentiment-analysis", tokenizer=tokenizer, model=model) |
|
|
|
def adjust(x): |
|
if x<0: |
|
return 2*x+1 |
|
return 2*x-1 |
|
|
|
def sa2(s): |
|
res= sa(s) |
|
return [adjust(-1*r['score']) if r['label']=='negative' else adjust(r['score']) for r in res ] |
|
|
|
|
|
def get_examples(): |
|
|
|
return ["Bu filmi beğenmedim\n bu filmi beğendim\n ceketin çok güzel\n bugün ne yesek"] |
|
|
|
import pandas as pd |
|
|
|
import matplotlib.pyplot as plt |
|
def grfunc(comments): |
|
df=pd.DataFrame() |
|
c2=[s.strip() for s in comments.split("\n") if len(s.split())>2] |
|
df["scores"]= sa2(c2) |
|
df.plot(kind='hist') |
|
return plt.gcf() |
|
|
|
import gradio as gr |
|
|
|
iface = gr.Interface( |
|
fn=grfunc, |
|
inputs=gr.inputs.Textbox(placeholder="put your sentences line by line", lines=5), |
|
outputs="plot", |
|
examples=get_examples()) |
|
iface.launch() |
|
|
