Upload folder using huggingface_hub

.ipynb_checkpoints/agents-checkpoint.py

@@ -0,0 +1,56 @@
+from crewai import Agent
+# from langchain_community.tools import DuckDuckGoSearchRun
+from crewai_tools import tool, SerperDevTool
+from langchain_community.llms import HuggingFaceEndpoint
+
+from tools.crypt_news_tool import GetCryptoNews 
+from tools.search_duckduckgo_tool import SearchDuckDuckGo 
+# from tools.crypt_news_tool import GetCryptoNews 
+
+
+search_tool = SerperDevTool()
+
+repo_id = "mistralai/Mistral-7B-Instruct-v0.2"
+llm = HuggingFaceEndpoint(
+    repo_id = repo_id, 
+    # huggingfacehub_api_token=HUGGINGFACEHUB_API_TOKEN,
+)
+
+class CryptoAnalysisAgents():
+    def crypto_analyst(self):
+        return Agent(
+            role='Crypto Analyst',
+            goal='Scrape the latest cryptocurrency news, indicators, tweets and current price of the target coin.',
+            backstory=(
+                'Keen on staying ahead with the most current information, '
+                'the Crypto Analyst scours through various media to gather the latest news and social sentiments.'
+            ),
+            tools=[
+                search_tool, 
+                # GetCryptoNews(), 
+                SearchDuckDuckGo()
+            ],
+            allow_delegation=False,
+            llm=llm,
+            verbose=True,
+            cache=True, 
+            max_iter=20, 
+        )
+
+    def content_writer(self):
+        return Agent(
+            role='Content Writer',
+            goal='Synthesize data and write engaging articles on latest trend of the target coin',
+            backstory=(
+                'Armed with a flair for storytelling, the Content Writer transforms intricate crypto data into captivating narratives,'
+                'illuminating the latest trends for a wide audience.'
+            ),
+            tools=[search_tool, 
+                   # AskWolframAlpha(),
+                  ],
+            allow_delegation=False,
+            llm=llm,
+            verbose=True,
+            cache=True, 
+            max_iter=20,  
+        )

.ipynb_checkpoints/chat_gradio-checkpoint.py

@@ -0,0 +1,44 @@
+from langchain_community.llms import HuggingFaceEndpoint
+from langchain.prompts import PromptTemplate
+from langchain.schema import AIMessage, HumanMessage
+from langchain.chains import LLMChain
+import gradio as gr
+import os
+
+from dotenv import load_dotenv
+load_dotenv()
+
+repo_id = "mistralai/Mistral-7B-Instruct-v0.2"
+llm = HuggingFaceEndpoint(
+    repo_id = repo_id, 
+)
+
+template = """You're a good chatbot. Answer this request: {question}
+Answer: Let's think step by step."""
+prompt = PromptTemplate.from_template(template=template)
+llm_chain = LLMChain(llm=llm, prompt=prompt)
+
+
+def predict(message, history):
+    history_langchain_format = []
+    # for human, ai in history:
+    #     history_langchain_format.append(HumanMessage(content=human))
+    #     history_langchain_format.append(AIMessage(content=ai))
+    # history_langchain_format.append(HumanMessage(content=message))
+    # gpt_response = llm(history_langchain_format)
+    response = llm_chain.invoke(message)['text']
+    return response
+
+gr.ChatInterface(predict).launch()
+
+
+
+
+
+
+
+
+
+
+
+

.ipynb_checkpoints/crew-checkpoint.py

@@ -0,0 +1,64 @@
+from dotenv import load_dotenv, find_dotenv
+from textwrap import dedent
+
+load_dotenv(find_dotenv())
+
+from agents import CryptoAnalysisAgents
+from tasks import CryptoAnalysisTasks
+from crewai import Crew
+
+
+class CryptoCrew:
+    def __init__(self, coin):
+        self.coin = coin
+
+    def run(self):
+        agents = CryptoAnalysisAgents()
+        tasks = CryptoAnalysisTasks()
+
+        crypto_analyst = agents.crypto_analyst()
+        content_writer = agents.content_writer()
+
+        research_task = tasks.research(crypto_analyst, self.coin)
+        recommend_task = tasks.recommend(content_writer, self.coin)
+
+        crew = Crew(
+            agents=[
+                crypto_analyst,
+                content_writer,
+            ],
+            tasks=[
+                research_task,
+                recommend_task,
+            ],
+            verbose=True
+        )
+
+        result = crew.kickoff()
+        return result
+
+# if __name__ == "__main__":
+#     print("## Welcome to Crypto Analysis Crew")
+#     print('-------------------------------')
+#     company = input(
+#         dedent("""
+#             Which cryptocurrency are you looking to delve into?
+#     """))
+
+#     crypto_crew = CryptoCrew(company)
+#     result = crypto_crew.run()
+#     print("\n\n########################")
+#     print("## Here is the Report")
+#     print("########################\n")
+    
+# print(result)
+
+
+
+
+
+
+
+
+
+

.ipynb_checkpoints/requirements-checkpoint.txt

@@ -0,0 +1,13 @@
+langchain 
+crewai 
+wolframalpha 
+duckduckgo-search
+crewai[tools]
+beautifulsoup4 
+requests 
+lxml
+langchain-community
+huggingface_hub
+google-api-python-client 
+google-auth-httplib2 
+google-auth-oauthlib

.ipynb_checkpoints/run_local-checkpoint.py

@@ -0,0 +1,61 @@
+# from langchain_community.llms import HuggingFaceEndpoint
+# from langchain.prompts import PromptTemplate
+# from langchain.schema import AIMessage, HumanMessage
+# from langchain.chains import LLMChain
+import gradio as gr
+import os
+
+from crew import CryptoCrew
+
+from dotenv import load_dotenv
+
+load_dotenv()
+
+def predict(message):
+    # company = input(
+    #     dedent("""
+    #         Which cryptocurrency are you looking to delve into?
+    # """))
+    # history_langchain_format = []
+    # for human, ai in history:
+    #     history_langchain_format.append(HumanMessage(content=human))
+    #     history_langchain_format.append(AIMessage(content=ai))
+    # history_langchain_format.append(HumanMessage(content=message))
+    # gpt_response = llm(history_langchain_format)
+    # response = llm_chain.invoke(message)['text']
+
+    crypto_crew = CryptoCrew(company)
+    response = "## Here is the Report\n\n" + crypto_crew.run()
+    
+    return response
+
+gr.ChatInterface(predict).launch()
+
+# if __name__ == "__main__":
+#     print("## Welcome to Crypto Analysis Crew")
+#     print('-------------------------------')
+#     company = input(
+#         dedent("""
+#             Which cryptocurrency are you looking to delve into?
+#     """))
+
+#     crypto_crew = CryptoCrew(company)
+#     result = crypto_crew.run()
+#     print("\n\n########################")
+#     print("## Here is the Report")
+#     print("########################\n")
+    
+# print(result)
+
+
+
+
+
+
+
+
+
+
+
+
+

.ipynb_checkpoints/tasks-checkpoint.py

@@ -0,0 +1,26 @@
+from crewai import Task
+from textwrap import dedent
+
+class CryptoAnalysisTasks():
+    def research(self, agent, coin):
+        return Task(description=dedent(f"""
+                                  Conduct a comprehensive analysis of the crypto coin {coin} .
+                                  Make sure to use the most recent data as possible.
+                                  Identify reliably potential oppotunity of the coin: {coin}.
+                                  """),
+                               expected_output="Full analysis report in bullet points",
+                               # tools=[GetCryptoNews()],
+                               agent=agent,
+                              )
+
+    def recommend(self, agent, coin):
+        return Task(description=dedent(f"""
+                                            Review and synthesize the analyses provided by the Crypto Analyst.
+                                            Combine these insights to form a comprehensive investment recommendation.
+                                            Answer if it's worth to invest in the coin {coin}. """),
+                    expected_output='A article of recommendations in bullet points',
+                    agent=agent,
+                   )
+
+    def __tip_section(self):
+        return "If you do your BEST WORK, I'll give you a $10,000 commission!"

README.md

@@ -1,12 +1,6 @@
 ---
-title: Crypto CrewAI
-emoji: 😻
-colorFrom: blue
-colorTo: indigo
+title: crypto_crewAI
+app_file: chat_gradio.py
 sdk: gradio
 sdk_version: 4.26.0
-app_file: app.py
-pinned: false
 ---
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

agents.py

@@ -0,0 +1,56 @@
+from crewai import Agent
+# from langchain_community.tools import DuckDuckGoSearchRun
+from crewai_tools import tool, SerperDevTool
+from langchain_community.llms import HuggingFaceEndpoint
+
+from tools.crypt_news_tool import GetCryptoNews 
+from tools.search_duckduckgo_tool import SearchDuckDuckGo 
+# from tools.crypt_news_tool import GetCryptoNews 
+
+
+search_tool = SerperDevTool()
+
+repo_id = "mistralai/Mistral-7B-Instruct-v0.2"
+llm = HuggingFaceEndpoint(
+    repo_id = repo_id, 
+    # huggingfacehub_api_token=HUGGINGFACEHUB_API_TOKEN,
+)
+
+class CryptoAnalysisAgents():
+    def crypto_analyst(self):
+        return Agent(
+            role='Crypto Analyst',
+            goal='Scrape the latest cryptocurrency news, indicators, tweets and current price of the target coin.',
+            backstory=(
+                'Keen on staying ahead with the most current information, '
+                'the Crypto Analyst scours through various media to gather the latest news and social sentiments.'
+            ),
+            tools=[
+                search_tool, 
+                # GetCryptoNews(), 
+                SearchDuckDuckGo()
+            ],
+            allow_delegation=False,
+            llm=llm,
+            verbose=True,
+            cache=True, 
+            max_iter=20, 
+        )
+
+    def content_writer(self):
+        return Agent(
+            role='Content Writer',
+            goal='Synthesize data and write engaging articles on latest trend of the target coin',
+            backstory=(
+                'Armed with a flair for storytelling, the Content Writer transforms intricate crypto data into captivating narratives,'
+                'illuminating the latest trends for a wide audience.'
+            ),
+            tools=[search_tool, 
+                   # AskWolframAlpha(),
+                  ],
+            allow_delegation=False,
+            llm=llm,
+            verbose=True,
+            cache=True, 
+            max_iter=20,  
+        )

chat_gradio.py

@@ -0,0 +1,44 @@
+from langchain_community.llms import HuggingFaceEndpoint
+from langchain.prompts import PromptTemplate
+from langchain.schema import AIMessage, HumanMessage
+from langchain.chains import LLMChain
+import gradio as gr
+import os
+
+from dotenv import load_dotenv
+load_dotenv()
+
+repo_id = "mistralai/Mistral-7B-Instruct-v0.2"
+llm = HuggingFaceEndpoint(
+    repo_id = repo_id, 
+)
+
+template = """You're a good chatbot. Answer this request: {question}
+Answer: Let's think step by step."""
+prompt = PromptTemplate.from_template(template=template)
+llm_chain = LLMChain(llm=llm, prompt=prompt)
+
+
+def predict(message, history):
+    history_langchain_format = []
+    # for human, ai in history:
+    #     history_langchain_format.append(HumanMessage(content=human))
+    #     history_langchain_format.append(AIMessage(content=ai))
+    # history_langchain_format.append(HumanMessage(content=message))
+    # gpt_response = llm(history_langchain_format)
+    response = llm_chain.invoke(message)['text']
+    return response
+
+gr.ChatInterface(predict).launch()
+
+
+
+
+
+
+
+
+
+
+
+

crew.py

@@ -0,0 +1,64 @@
+from dotenv import load_dotenv, find_dotenv
+from textwrap import dedent
+
+load_dotenv(find_dotenv())
+
+from agents import CryptoAnalysisAgents
+from tasks import CryptoAnalysisTasks
+from crewai import Crew
+
+
+class CryptoCrew:
+    def __init__(self, coin):
+        self.coin = coin
+
+    def run(self):
+        agents = CryptoAnalysisAgents()
+        tasks = CryptoAnalysisTasks()
+
+        crypto_analyst = agents.crypto_analyst()
+        content_writer = agents.content_writer()
+
+        research_task = tasks.research(crypto_analyst, self.coin)
+        recommend_task = tasks.recommend(content_writer, self.coin)
+
+        crew = Crew(
+            agents=[
+                crypto_analyst,
+                content_writer,
+            ],
+            tasks=[
+                research_task,
+                recommend_task,
+            ],
+            verbose=True
+        )
+
+        result = crew.kickoff()
+        return result
+
+# if __name__ == "__main__":
+#     print("## Welcome to Crypto Analysis Crew")
+#     print('-------------------------------')
+#     company = input(
+#         dedent("""
+#             Which cryptocurrency are you looking to delve into?
+#     """))
+
+#     crypto_crew = CryptoCrew(company)
+#     result = crypto_crew.run()
+#     print("\n\n########################")
+#     print("## Here is the Report")
+#     print("########################\n")
+    
+# print(result)
+
+
+
+
+
+
+
+
+
+

requirements.txt

@@ -0,0 +1,13 @@
+langchain 
+crewai 
+wolframalpha 
+duckduckgo-search
+crewai[tools]
+beautifulsoup4 
+requests 
+lxml
+langchain-community
+huggingface_hub
+google-api-python-client 
+google-auth-httplib2 
+google-auth-oauthlib

run_local.py

@@ -0,0 +1,61 @@
+# from langchain_community.llms import HuggingFaceEndpoint
+# from langchain.prompts import PromptTemplate
+# from langchain.schema import AIMessage, HumanMessage
+# from langchain.chains import LLMChain
+import gradio as gr
+import os
+
+from crew import CryptoCrew
+
+from dotenv import load_dotenv
+
+load_dotenv()
+
+def predict(message):
+    # company = input(
+    #     dedent("""
+    #         Which cryptocurrency are you looking to delve into?
+    # """))
+    # history_langchain_format = []
+    # for human, ai in history:
+    #     history_langchain_format.append(HumanMessage(content=human))
+    #     history_langchain_format.append(AIMessage(content=ai))
+    # history_langchain_format.append(HumanMessage(content=message))
+    # gpt_response = llm(history_langchain_format)
+    # response = llm_chain.invoke(message)['text']
+
+    crypto_crew = CryptoCrew(company)
+    response = "## Here is the Report\n\n" + crypto_crew.run()
+    
+    return response
+
+gr.ChatInterface(predict).launch()
+
+# if __name__ == "__main__":
+#     print("## Welcome to Crypto Analysis Crew")
+#     print('-------------------------------')
+#     company = input(
+#         dedent("""
+#             Which cryptocurrency are you looking to delve into?
+#     """))
+
+#     crypto_crew = CryptoCrew(company)
+#     result = crypto_crew.run()
+#     print("\n\n########################")
+#     print("## Here is the Report")
+#     print("########################\n")
+    
+# print(result)
+
+
+
+
+
+
+
+
+
+
+
+
+

tasks.py

@@ -0,0 +1,26 @@
+from crewai import Task
+from textwrap import dedent
+
+class CryptoAnalysisTasks():
+    def research(self, agent, coin):
+        return Task(description=dedent(f"""
+                                  Conduct a comprehensive analysis of the crypto coin {coin} .
+                                  Make sure to use the most recent data as possible.
+                                  Identify reliably potential oppotunity of the coin: {coin}.
+                                  """),
+                               expected_output="Full analysis report in bullet points",
+                               # tools=[GetCryptoNews()],
+                               agent=agent,
+                              )
+
+    def recommend(self, agent, coin):
+        return Task(description=dedent(f"""
+                                            Review and synthesize the analyses provided by the Crypto Analyst.
+                                            Combine these insights to form a comprehensive investment recommendation.
+                                            Answer if it's worth to invest in the coin {coin}. """),
+                    expected_output='A article of recommendations in bullet points',
+                    agent=agent,
+                   )
+
+    def __tip_section(self):
+        return "If you do your BEST WORK, I'll give you a $10,000 commission!"

tools/.ipynb_checkpoints/crypt_news_tool-checkpoint.py

@@ -0,0 +1,30 @@
+import requests
+from bs4 import BeautifulSoup
+from crewai_tools import BaseTool
+
+class GetCryptoNews(BaseTool):
+    name: str = "GetCryptoNews"
+    description: str = "fetch news from https://cryptonews.com/"
+
+    def _run(self, argument: str) -> str:
+        url = "https://cryptonews.com/"
+        response = requests.get(url)
+        webpage = response.text
+        soup = BeautifulSoup(webpage,'html.parser')
+        articles = soup.select("#main-news-left .title-news-slider")
+
+        data = ""
+        for article in articles[:5]: 
+            title = article.text.strip() 
+            url = article['href']  
+        
+            response = requests.get(url)
+            article_page = response.text
+            soup = BeautifulSoup(article_page, 'html.parser')
+            
+            data += f'Title: {title}, Content: '
+            
+            for section in soup.find_all(['h1', 'h2', 'h3', 'p']):
+                data += section.text.strip()
+                
+        return data

tools/.ipynb_checkpoints/search_duckduckgo_tool-checkpoint.py

@@ -0,0 +1,11 @@
+from crewai_tools import BaseTool
+from langchain_community.tools import DuckDuckGoSearchRun
+
+
+class SearchDuckDuckGo(BaseTool):
+    name: str = "SearchDuckDuckGo"
+    description: str = "use DuckDuckGo to fetch some news"
+    def _run(self, question: str) -> str:
+        search = DuckDuckGoSearchRun()
+        result = search.run(question)
+        return result

tools/.ipynb_checkpoints/wolfram_alpha_tool-checkpoint.py

@@ -0,0 +1,17 @@
+from crewai_tools import BaseTool
+
+from langchain_community.utilities.wolfram_alpha import WolframAlphaAPIWrapper
+
+wolfram = WolframAlphaAPIWrapper()
+
+class AskWolframAlpha(BaseTool):
+    name: str = "AskWolframAlpha"
+    description: str = "use WolframAlpha to analyze data"   
+
+    def _run(self, question: str) -> str:
+        "Ask Wolfram Alpha to do math, e.g : descriptive statistics {20.9,23.2,26.2,26.4}" 
+        wolfram = WolframAlphaAPIWrapper()
+        return wolfram.run(question)
+
+# Test
+# print(AskWolframAlpha()._run("descriptive statistics {20.9,23.2,26.2,26.4,20.9,23.2,26.2,26.4,20.9}"))

tools/crypt_news_tool.py

@@ -0,0 +1,30 @@
+import requests
+from bs4 import BeautifulSoup
+from crewai_tools import BaseTool
+
+class GetCryptoNews(BaseTool):
+    name: str = "GetCryptoNews"
+    description: str = "fetch news from https://cryptonews.com/"
+
+    def _run(self, argument: str) -> str:
+        url = "https://cryptonews.com/"
+        response = requests.get(url)
+        webpage = response.text
+        soup = BeautifulSoup(webpage,'html.parser')
+        articles = soup.select("#main-news-left .title-news-slider")
+
+        data = ""
+        for article in articles[:5]: 
+            title = article.text.strip() 
+            url = article['href']  
+        
+            response = requests.get(url)
+            article_page = response.text
+            soup = BeautifulSoup(article_page, 'html.parser')
+            
+            data += f'Title: {title}, Content: '
+            
+            for section in soup.find_all(['h1', 'h2', 'h3', 'p']):
+                data += section.text.strip()
+                
+        return data

tools/search_duckduckgo_tool.py

@@ -0,0 +1,11 @@
+from crewai_tools import BaseTool
+from langchain_community.tools import DuckDuckGoSearchRun
+
+
+class SearchDuckDuckGo(BaseTool):
+    name: str = "SearchDuckDuckGo"
+    description: str = "use DuckDuckGo to fetch some news"
+    def _run(self, question: str) -> str:
+        search = DuckDuckGoSearchRun()
+        result = search.run(question)
+        return result

tools/wolfram_alpha_tool.py

@@ -0,0 +1,17 @@
+from crewai_tools import BaseTool
+
+from langchain_community.utilities.wolfram_alpha import WolframAlphaAPIWrapper
+
+wolfram = WolframAlphaAPIWrapper()
+
+class AskWolframAlpha(BaseTool):
+    name: str = "AskWolframAlpha"
+    description: str = "use WolframAlpha to analyze data"   
+
+    def _run(self, question: str) -> str:
+        "Ask Wolfram Alpha to do math, e.g : descriptive statistics {20.9,23.2,26.2,26.4}" 
+        wolfram = WolframAlphaAPIWrapper()
+        return wolfram.run(question)
+
+# Test
+# print(AskWolframAlpha()._run("descriptive statistics {20.9,23.2,26.2,26.4,20.9,23.2,26.2,26.4,20.9}"))