prompt changes

app.py

@@ -14,10 +14,15 @@ system_template = """\
 ###Instruction###
 You are an expert assistant answering technical questions on machine learning and deep learning subject.
 Ensure that your response is unbiased and generic, you will be 'AWARDED' for giving really good clarity and correct answers.
+##EXAMPLES##
+If users asks 'explain neural networks', your response should be with an overview of neural networks, discussing how they are computational models inspired by the human brain that are used to recognize patterns and solve complex problems in machine learning.
+If users ask 'code convolutional neural network', your response should contain example of the code necessary to create a CNN.
+If users ask 'resource grdient descent', your response should offer links to tutorials, video lectures, or articles that explain gradient descent, which is an optimization algorithm used to minimize a function by iteratively moving in the direction of steepest descent.
+If users ask 'project sentiment analysis', your response should discuss the steps involved in creating a sentiment analysis model, such as data collection, preprocessing, model selection, training, and evaluation, and potentially offer advice on best practices or methodologies to consider.
 """
 
-user_template = """{input} + 
-Think and give either code, links or steps for the question asked along with response.
+user_template = """{input} \n + 
+Think and give either explanation or code or links for resources or steps for project, for the questions asked along with response.
 """
 
 

chainlit.md

@@ -17,8 +17,8 @@ This Chainlit app was created following instructions from [this repository!](htt
 **Ready to dive in? Here are some commands to get you started:**
 
 - `explain <neural networks>`: A user looking for an understanding of neural networks might use this command. The chatbot could respond with an overview of neural networks, discussing how they are computational models inspired by the human brain that are used to recognize patterns and solve complex problems in machine learning.
-- `code convolutional neural network`: If a user wants to see how to implement a convolutional neural network in Python using a library like TensorFlow or PyTorch, this command could trigger the chatbot to provide a basic example of the code necessary to create a CNN for image classification.
-- `resource gradient descent`: When a user needs educational resources to learn about gradient descent, this command might lead the chatbot to offer links to tutorials, video lectures, or articles that explain gradient descent, which is an optimization algorithm used to minimize a function by iteratively moving in the direction of steepest descent.
-- `project sentiment analysis`: If a user is starting a project on sentiment analysis, this command could prompt the chatbot to discuss the steps involved in creating a sentiment analysis model, such as data collection, preprocessing, model selection, training, and evaluation, and potentially offer advice on best practices or methodologies to consider.
+- `code <convolutional neural network>`: If a user wants to see how to implement a convolutional neural network in Python using a library like TensorFlow or PyTorch, this command could trigger the chatbot to provide a basic example of the code necessary to create a CNN for image classification.
+- `<resource gradient descent>`: When a user needs educational resources to learn about gradient descent, this command might lead the chatbot to offer links to tutorials, video lectures, or articles that explain gradient descent, which is an optimization algorithm used to minimize a function by iteratively moving in the direction of steepest descent.
+- `<project sentiment analysis>`: If a user is starting a project on sentiment analysis, this command could prompt the chatbot to discuss the steps involved in creating a sentiment analysis model, such as data collection, preprocessing, model selection, training, and evaluation, and potentially offer advice on best practices or methodologies to consider.
 
 **Let's start learning and coding together!**