Update install

docs/dspy.md

@@ -6,7 +6,7 @@ sidebar_position: 2
 
 ## Programming not Prompting
 
-**[DSPy](https://dspy-docs.vercel.app/)** is a machine learning framework for building LLM applications with a systematic approach by abstracting prompt engineering techniques and offering ways to evaluate your pipelines. Often when tackling a difficult task you will have to provide examples to your LLM, beside being time-consuming, handcrafting these examples can lead to the brittleness of your application specially if it involve numerous calls to LMs (like HybridAGI). When you change the LMs, you will have to re-design the prompts for that model. DSPy change that by emphazing small multi-input/multi-output prompts with automatic building of examples allowing a better control.
+**[DSPy](https://dspy-docs.vercel.app/)** is a machine learning framework for building LLM applications with a systematic approach by abstracting prompt engineering techniques and offering ways to evaluate your pipelines. Often when tackling a difficult task you will have to provide examples to your LLM, beside being time-consuming, handcrafting these examples can lead to the brittleness of your application specially if it involve numerous calls to LMs (like HybridAGI). When you change the pipeline, you will have to re-design the examples. DSPy change that by emphazing small multi-input/multi-output prompts with automatic building of examples allowing a better control and faster iteration.
 
 ## DSPy language models
 
@@ -148,7 +148,7 @@ class CheckHelpfullnessSignature(dspy.Signature):
         prefix="Helpful[Yes/No:]",
     )
 
-def check_helpfulness(example, prediction):
+def check_helpfulness(example, prediction, teacher_lm = Optional[dspy.LM]):
     # This line means that we discard the example if the agent reached the max iterations
     # Meaning it was probably stuck in a loop
     if prediction.finish_reason == "max iters":

docs/install.md

@@ -10,36 +10,21 @@ sidebar_position: 4
 
 Additionally you can install [Ollama](https://ollama.com/) to run the examples, pull `mistral` model and start the ollama server.
 
-### 1. Install
-<!-- 
-#### From pip
-
-```
-virtualenv venv
-source venv/bin/activate
-pip install git+https://github.com/SynaLinks/HybridAGI
-```
-
-#### From sources -->
+### 1. Install using pip
 
 ```bash
-git clone https://github.com/SynaLinks/HybridAGI
-cd HybridAGI
-virtualenv venv
-source venv/bin/activate
-pip install poetry # If you don't have it already
-poetry install
+pip install hybridagi
 ```
 
+Note: Before installing `hybridagi` we recommand you to create a virtual environment using virtualenv, conda or your preferred environment manager.
+
 ### 2. Setup the Knowledge Base (needed for the system to work)
 
-<!-- ```
-docker run -p 6379:6379 -it --rm falkordb/falkordb:edge
-``` -->
+Then setup the knowledge base using docker:
+
 ```bash
-docker compose up
+docker run -p 6379:6379 -p 3000:3000 -it --rm falkordb/falkordb:edge
 ```
-
 ### 3. Start programmming your Agent
 
 You can start to learn more about graph prompt programming in this **[first tutorial](tutorials/basics/graph-prompt-programming.md)** or with this **[Python notebook](https://github.com/SynaLinks/HybridAGI/blob/main/notebooks/first_steps.ipynb)**.