2

app.py

@@ -2,7 +2,7 @@ import streamlit as st
 from widgets import *
 
 
-# [![github](https://img.kookapp.cn/assets/2022-09/1w4G0FIWGK00w00w.png)](https://github.com/Mondkuchen/idp_LiteratureResearch_Tool)
+
 
 # sidebar content
 platforms, number_papers, start_year, end_year, hyperparams = render_sidebar()
@@ -29,16 +29,7 @@ if submitted:
                 show_preview, start_year, end_year,
                 hyperparams,
                 hyperparams['standardization'])
-    # '''
-    # bar = (
-    #     Bar()
-    #     .add_xaxis(["Cluster 1", "Cluster 2", "Cluster 3", 'Cluster 4', 'Cluster 5'])
-    #     .add_yaxis("numbers", [23, 16, 13, 12, 5])
-    #     .set_global_opts(title_opts=opts.TitleOpts(title="Fake Data"))
-    # )
-    #
-    # components.html(generate_html_pyecharts(bar, 'tmp.html'), height=500, width=1000)
-    # '''
+
 
 
 

lrt/utils/functions.py

@@ -7,6 +7,7 @@ from transformers import AutoTokenizer, AutoModelForSeq2SeqLM,Text2TextGeneratio
 from inference_hf import InferenceHF
 from .dimension_reduction import PCA
 from unsupervised_learning.clustering import GaussianMixture
+from models import KeyBartAdapter
 
 class Template:
     def __init__(self):
@@ -114,7 +115,7 @@ def __create_model__(model_ckpt):
 
         return ret
 
-    elif model_ckpt == 'KeyBartAdapter' or model_ckpt == 'KeyBart':
+    elif model_ckpt == 'KeyBart':
         model_ckpt = template.keywords_extraction[model_ckpt]
         def ret(texts: List[str]):
             # first try inference API
@@ -154,6 +155,26 @@ def __create_model__(model_ckpt):
                 return results
 
         return ret
+
+    elif model_ckpt == 'KeyBartAdapter':
+        def ret(texts: List[str]):
+            model = KeyBartAdapter.from_pretrained('Adapting/KeyBartAdapter',revision='3aee5ecf1703b9955ab0cd1b23208cc54eb17fce', adapter_hid_dim=32)
+            tokenizer = AutoTokenizer.from_pretrained("bloomberg/KeyBART")
+            pipe = Text2TextGenerationPipeline(model=model, tokenizer=tokenizer)
+
+            tmp = pipe(texts)
+            results = [
+                set(
+                    map(str.strip,
+                        x['generated_text'].split(';')  # [str...]
+                        )
+                )
+                for x in tmp]  # [{str...}...]
+
+            return results
+        return ret
+
+
     else:
         raise RuntimeError(f'The model {model_ckpt} is not supported. Please open an issue on the GitHub about the model.')
 

requirements.txt

@@ -11,4 +11,5 @@ transformers==4.22.1
 textdistance==4.5.0
 datasets==2.5.2
 bokeh==2.4.1
-ml-leoxiang66
+ml-leoxiang66
+KeyBartAdapter