Add Gradio implementation

app.py

@@ -0,0 +1,117 @@
+import gradio as gr
+import nltk
+import requests
+from bs4 import BeautifulSoup
+from fake_headers import Headers
+from newspaper import Article
+from nltk.tokenize import sent_tokenize
+from summa.summarizer import summarize
+from transformers import pipeline
+
+nltk.download("punkt")
+
+
+def get_news(stock, n_results):
+    header_gen = Headers(headers=True)
+    headers = header_gen.generate()
+
+    params = {"q": stock + " stock news"}
+
+    response = requests.get(
+        "https://news.google.com/search", headers=headers, params=params
+    )
+    soup = BeautifulSoup(response.text, "lxml")
+
+    url_elems = soup.find_all(class_="VDXfz")
+    urls = []
+
+    for elem in url_elems:
+        urls.append("https://news.google.com" + elem["href"][1:])
+
+    articles = []
+    total_results = 0
+
+    for url in urls:
+        if total_results == n_results:
+            break
+        try:
+            article = Article(url)
+            article.download()
+            article.parse()
+            article = article.text
+
+            if len(article.split()) > 50:
+                articles.append(article)
+                total_results += 1
+        except:
+            continue
+
+    return articles
+
+
+def gen_corpus(articles, stock, length):
+    raw_corpus = ""
+
+    for article in articles:
+        article_sent = sent_tokenize(article)
+
+        for sent in article_sent:
+            if len(sent.split()) > 5 and stock.lower() in sent.lower():
+                raw_corpus += sent
+
+    corpus = summarize(raw_corpus, words=length)
+    return corpus
+
+
+def gen_summary(corpus, length):
+    summarizer = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6")
+
+    min_length = int(0.8 * length)
+    max_length = int(1.2 * length)
+    raw_summary = summarizer(
+        corpus[:1024], min_length=min_length, max_length=max_length
+    )[0]["summary_text"]
+
+    summary_sent = sent_tokenize(raw_summary)
+    summary = ""
+
+    for sent in summary_sent:
+        summary += f"-{sent}\n"
+
+    return summary
+
+
+def gen_sentiment(corpus):
+    analyzer = pipeline("sentiment-analysis", model="ProsusAI/finbert")
+    raw_sentiment = analyzer(corpus[:512], return_all_scores=True)
+
+    sentiment = {}
+
+    for sentiment_dict in raw_sentiment[0]:
+        label = sentiment_dict["label"]
+        value = sentiment_dict["score"]
+
+        if label == "positive":
+            label = "bull"
+        elif label == "negative":
+            label = "bear"
+        else:
+            label = "neutral"
+
+        sentiment[label] = value
+
+    return sentiment
+
+
+def gen_report(stock, length):
+    news = get_news(stock, length / 5)
+    corpus = gen_corpus(news, stock, length * 3)
+    summary = gen_summary(corpus, length)
+    sentiment = gen_sentiment(corpus)
+
+    return summary, sentiment
+
+
+iface = gr.Interface(
+    fn=gen_report, inputs=["text", "number"], outputs=["text", "label"]
+).launch()

requirements.txt

@@ -0,0 +1,133 @@
+#
+# These requirements were autogenerated by pipenv
+# To regenerate from the project's Pipfile, run:
+#
+#    pipenv lock --requirements
+#
+
+-i https://pypi.org/simple
+absl-py==1.0.0; python_version >= '3.6'
+aiohttp==3.8.1; python_version >= '3.6'
+aiosignal==1.2.0; python_version >= '3.6'
+analytics-python==1.4.0
+anyio==3.5.0; python_full_version >= '3.6.2'
+asgiref==3.5.0; python_version >= '3.7'
+astunparse==1.6.3
+async-timeout==4.0.2; python_version >= '3.6'
+attrs==21.4.0; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3, 3.4'
+backoff==1.10.0; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3, 3.4'
+bcrypt==3.2.0; python_version >= '3.6'
+beautifulsoup4==4.10.0
+black==22.1.0
+bs4==0.0.1
+cachetools==5.0.0; python_version ~= '3.7'
+certifi==2021.10.8
+cffi==1.15.0
+charset-normalizer==2.0.12; python_version >= '3'
+click==8.0.4; python_version >= '3.6'
+cryptography==36.0.1; python_version >= '3.6'
+cssselect==1.1.0; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3'
+cycler==0.11.0; python_version >= '3.6'
+fake-headers==1.0.2
+fastapi==0.74.1; python_full_version >= '3.6.1'
+feedfinder2==0.0.4
+feedparser==6.0.8; python_version >= '3.6'
+ffmpy==0.3.0
+filelock==3.6.0; python_version >= '3.7'
+flatbuffers==2.0
+fonttools==4.29.1; python_version >= '3.7'
+frozenlist==1.3.0; python_version >= '3.7'
+gast==0.5.3; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3'
+google-auth-oauthlib==0.4.6; python_version >= '3.6'
+google-auth==2.6.0; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3, 3.4, 3.5'
+google-pasta==0.2.0
+gradio==2.8.5
+grpcio==1.44.0; python_version >= '3.6'
+h11==0.13.0; python_version >= '3.6'
+h5py==3.6.0; python_version >= '3.7'
+html5lib==1.1; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3, 3.4'
+huggingface-hub==0.4.0; python_version >= '3.6'
+idna==3.3; python_version >= '3'
+importlib-metadata==4.11.2; python_version < '3.10'
+isort==5.10.1
+jieba3k==0.35.1
+jinja2==3.0.3
+joblib==1.1.0; python_version >= '3.6'
+keras-preprocessing==1.1.2
+keras==2.8.0
+kiwisolver==1.3.2; python_version >= '3.7'
+libclang==13.0.0
+linkify-it-py==1.0.3
+lxml==4.8.0; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3, 3.4'
+markdown-it-py[linkify,plugins]==2.0.1; python_version ~= '3.6'
+markdown==3.3.6; python_version >= '3.6'
+markupsafe==2.1.0; python_version >= '3.7'
+matplotlib==3.5.1; python_version >= '3.7'
+mdit-py-plugins==0.3.0
+mdurl==0.1.0; python_version >= '3.6'
+monotonic==1.6
+multidict==6.0.2; python_version >= '3.7'
+mypy-extensions==0.4.3
+newspaper3k==0.2.8
+nltk==3.7
+numpy==1.22.2; python_version >= '3.8'
+oauthlib==3.2.0; python_version >= '3.6'
+opt-einsum==3.3.0; python_version >= '3.5'
+orjson==3.6.7; python_version >= '3.7'
+packaging==21.3; python_version >= '3.6'
+pandas==1.4.1; python_version >= '3.8'
+paramiko==2.9.2
+pathspec==0.9.0
+pillow==9.0.1; python_version >= '3.7'
+platformdirs==2.5.1; python_version >= '3.7'
+protobuf==3.19.4; python_version >= '3.5'
+pyasn1-modules==0.2.8
+pyasn1==0.4.8
+pycparser==2.21
+pycryptodome==3.14.1; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3, 3.4'
+pydantic==1.9.0; python_full_version >= '3.6.1'
+pydub==0.25.1
+pynacl==1.5.0; python_version >= '3.6'
+pyparsing==3.0.7; python_version >= '3.6'
+python-dateutil==2.8.2; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3'
+python-multipart==0.0.5
+pytz==2021.3
+pyyaml==6.0; python_version >= '3.6'
+regex==2022.1.18
+requests-file==1.5.1
+requests-oauthlib==1.3.1; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3'
+requests==2.27.1
+rsa==4.8; python_version >= '3.6'
+sacremoses==0.0.47
+scipy==1.8.0; python_version < '3.11' and python_version >= '3.8'
+setuptools==60.9.3; python_version >= '3.7'
+sgmllib3k==1.0.0
+six==1.16.0; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3'
+sniffio==1.2.0; python_version >= '3.5'
+soupsieve==2.3.1; python_version >= '3.6'
+starlette==0.17.1; python_version >= '3.6'
+summa==1.2.0
+tensorboard-data-server==0.6.1; python_version >= '3.6'
+tensorboard-plugin-wit==1.8.1
+tensorboard==2.8.0; python_version >= '3.6'
+tensorflow-io-gcs-filesystem==0.24.0; python_version < '3.11' and python_version >= '3.7'
+tensorflow==2.8.0
+termcolor==1.1.0
+tf-estimator-nightly==2.8.0.dev2021122109
+tinysegmenter==0.3
+tldextract==3.2.0; python_version >= '3.7'
+tokenizers==0.11.5
+tomli==2.0.1; python_version >= '3.7'
+torch==1.10.2
+tqdm==4.62.3; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3'
+transformers==4.16.2
+typing-extensions==4.1.1; python_version < '3.10'
+uc-micro-py==1.0.1; python_version >= '3.6'
+urllib3==1.26.8; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3, 3.4' and python_version < '4'
+uvicorn==0.17.5; python_version >= '3.7'
+webencodings==0.5.1
+werkzeug==2.0.3; python_version >= '3.6'
+wheel==0.37.1; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3, 3.4'
+wrapt==1.13.3; python_version >= '2.7' and python_version not in '3.0, 3.1, 3.2, 3.3, 3.4'
+yarl==1.7.2; python_version >= '3.6'
+zipp==3.7.0; python_version >= '3.7'