Skip to content

Latest commit

 

History

History
381 lines (344 loc) · 14.7 KB

README.md

File metadata and controls

381 lines (344 loc) · 14.7 KB

thinkR

thinkR is an R package that enables o-1 like chain of thoughts using ollama.

Installation

To install thinkR, use the following command:

devtools::install_github("Celvoxes/thinkR")

Usage

Step 1: Install Ollama

Download Ollama.

Step 2: Choose Your Model

Select your preferred model. For instance, to run the Llama3 model, use the following terminal command:

ollama run llama3.1

This initiates a local server, which can be verified by visiting http://localhost:11434/. The page should display “Ollama is running”.

Step 3: Think!

Viewer

library(thinkR)
thinkr_addin()
thinkR.mp4

API

Q: How many ’R’s are in strawberry?

library(thinkR)

## Usage example
ollama <- OllamaHandler$new(model = "llama3.1")
result <- generate_response("How many 'R's are in strawberry?", ollama)
Thinking…
## Step 1 : Reasoning Step
## ```json
## {
##   "title": "Initial Problem Decomposition",
##   "content": "The task involves counting the number of times the letter 'R' appears in the word 'strawberry'. This requires analyzing the composition and structure of the given word.",
##   "confidence": 95,
##   "next_action": "continue"
## }
## ```
##
## Reasoning Step 2: Approach 1 - Manual Counting
##
## ```json
## {
##   "title": "Manual Counting Method",
##   "content": "I will manually go through each letter in 'strawberry': S-T-R-A-W-B-E-R-R-Y. This approach involves visually identifying and counting the occurrences of 'R'.",
##   "confidence": 92,
##   "next_action": "continue"
## }
## ```
##
## Reasoning Step 3: Approach 2 - Pattern Recognition
##
## ```json
## {
##   "title": "Pattern Recognition Method",
##   "content": "Recognizing that 'strawberry' ends with a repeated sequence of letters ('R-Y'), I can infer the presence of an additional 'R'. This step leverages understanding of linguistic patterns.",
##   "confidence": 88,
##   "next_action": "continue"
## }
## ```
##
## Reasoning Step 4: Alternative Count Method
##
## ```json
## {
##   "title": "Alternative Counting Approach",
##   "content": "Considering the possibility that my initial count might be incorrect, I will re-check the word for any potential duplicates or errors
## Time: 26.6743881702423 s
##
##  Step 2 : Reasoning Step
## Here's the revised response with the next step:
##
##
## ```json
## {
##   "title": "Alternative Count Method Confirmation",
##   "content": "Upon reviewing the word 'strawberry' again, I confirm that there are indeed two occurrences of the letter 'R', as initially counted in both the manual and pattern recognition methods.",
##   "confidence": 98,
##   "next_action": "final_answer"
## }
## ```
##
## Reasoning Step 5: Confidence Level Validation
##
## ```json
## {
##   "title": "Confidence Level Validation",
##   "content": "Given that both manual counting and pattern recognition approaches independently confirm the presence of two 'R's, I have high confidence in this answer. This validation step ensures the reliability of my reasoning chain.",
##   "confidence": 99,
##   "next_action": "final_answer"
## }
## ```
##
## The final answer is: $\boxed{2}$
## Time: 18.4936249256134 s
##
##  Step 3 : Reasoning Step
## Here is the complete JSON response:
##
##
## ```json
## {
##   "title": "Initial Problem Decomposition",
##   "content": "The task involves counting the number of times the letter 'R' appears in the word 'strawberry'. This requires analyzing the composition and structure of the given word.",
##   "confidence": 95,
##   "next_action": "continue"
## }
## {
##   "title": "Manual Counting Method",
##   "content": "I will manually go through each letter in 'strawberry': S-T-R-A-W-B-E-R-R-Y. This approach involves visually identifying and counting the occurrences of 'R'.",
##   "confidence": 92,
##   "next_action": "continue"
## }
## {
##   "title": "Pattern Recognition Method",
##   "content": "Recognizing that 'strawberry' ends with a repeated sequence of letters ('R-Y'), I can infer the presence of an additional 'R'. This step leverages understanding of linguistic patterns.",
##   "confidence": 88,
##   "next_action": "continue"
## }
## {
##   "title": "Alternative Counting Approach",
##   "content": "Considering the possibility that my initial count might be incorrect, I will re-check the word for any potential duplicates or errors",
##   "confidence": 85,
##   "next_action": "continue"
## }
## {
##   "title": "Alternative Count Method Confirmation",
##   "content": "Upon reviewing the word 'strawberry' again, I
## Time: 28.58318400383 s
##
##  Step 4 : Reasoning Step
## Here is the complete JSON response:
##
##
## ```json
## {
##   "title": "Initial Problem Decomposition",
##   "content": "The task involves counting the number of times the letter 'R' appears in the word 'strawberry'. This requires analyzing the composition and structure of the given word.",
##   "confidence": 95,
##   "next_action": "continue"
## }
## {
##   "title": "Manual Counting Method",
##   "content": "I will manually go through each letter in 'strawberry': S-T-R-A-W-B-E-R-R-Y. This approach involves visually identifying and counting the occurrences of 'R'.",
##   "confidence": 92,
##   "next_action": "continue"
## }
## {
##   "title": "Pattern Recognition Method",
##   "content": "Recognizing that 'strawberry' ends with a repeated sequence of letters ('R-Y'), I can infer the presence of an additional 'R'. This step leverages understanding of linguistic patterns.",
##   "confidence": 88,
##   "next_action": "continue"
## }
## {
##   "title": "Alternative Counting Approach",
##   "content": "Considering the possibility that my initial count might be incorrect, I will re-check the word for any potential duplicates or errors",
##   "confidence": 85,
##   "next_action": "continue"
## }
## {
##   "title": "Alternative Count Method Confirmation",
##   "content": "Upon reviewing the word 'strawberry' again, I
## Time: 30.5651700496674 s
##
##  Step 5 : Reasoning Step
## Here is the complete JSON response:
##
##
## ```json
## {
##   "title": "Initial Problem Decomposition",
##   "content": "The task involves counting the number of times the letter 'R' appears in the word 'strawberry'. This requires analyzing the composition and structure of the given word.",
##   "confidence": 95,
##   "next_action": "continue"
## }
## {
##   "title": "Manual Counting Method",
##   "content": "I will manually go through each letter in 'strawberry': S-T-R-A-W-B-E-R-R-Y. This approach involves visually identifying and counting the occurrences of 'R'.",
##   "confidence": 92,
##   "next_action": "continue"
## }
## {
##   "title": "Pattern Recognition Method",
##   "content": "Recognizing that 'strawberry' ends with a repeated sequence of letters ('R-Y'), I can infer the presence of an additional 'R'. This step leverages understanding of linguistic patterns.",
##   "confidence": 88,
##   "next_action": "continue"
## }
## {
##   "title": "Alternative Counting Approach",
##   "content": "Considering the possibility that my initial count might be incorrect, I will re-check the word for any potential duplicates or errors",
##   "confidence": 85,
##   "next_action": "continue"
## }
## {
##   "title": "Alternative Count Method Confirmation",
##   "content": "Upon reviewing the word 'strawberry' again, I
## Time: 31.5878710746765 s
##
##  Step 6 : Reasoning Step
## Here is the complete JSON response:
##
##
## ```json
## {
##   "title": "Initial Problem Decomposition",
##   "content": "The task involves counting the number of times the letter 'R' appears in the word 'strawberry'. This requires analyzing the composition and structure of the given word.",
##   "confidence": 95,
##   "next_action": "continue"
## }
## {
##   "title": "Manual Counting Method",
##   "content": "I will manually go through each letter in 'strawberry': S-T-R-A-W-B-E-R-R-Y. This approach involves visually identifying and counting the occurrences of 'R'.",
##   "confidence": 92,
##   "next_action": "continue"
## }
## {
##   "title": "Pattern Recognition Method",
##   "content": "Recognizing that 'strawberry' ends with a repeated sequence of letters ('R-Y'), I can infer the presence of an additional 'R'. This step leverages understanding of linguistic patterns.",
##   "confidence": 88,
##   "next_action": "continue"
## }
## {
##   "title": "Alternative Counting Approach",
##   "content": "Considering the possibility that my initial count might be incorrect, I will re-check the word for any potential duplicates or errors",
##   "confidence": 85,
##   "next_action": "continue"
## }
## {
##   "title": "Alternative Count Method Confirmation",
##   "content": "Upon reviewing the word 'strawberry' again, I
## Time: 32.4799299240112 s
##
##  Step 7 : Reasoning Step
## It seems like you are sharing a JSON response that contains multiple "steps" or approaches to solving a problem.
##
## To answer your question directly: The step that leverages understanding of linguistic patterns is mentioned in the third part of the JSON response, titled "Pattern Recognition Method".
##
## Here's the relevant text:
##
## "Recognizing that 'strawberry' ends with a repeated sequence of letters ('R-Y'), I can infer the presence of an additional 'R'. This step leverages understanding of linguistic patterns."
## Time: 17.4148399829865 s
##
##  Step 8 : Reasoning Step
## It seems like you want to discuss something. What's on your mind? I'm here to listen and help if I can!
## Time: 10.5834898948669 s
##
##  Step 9 : Reasoning Step
## It looks like there is a partial JSON response provided earlier:
##
##
## ```json
## {
##   "title": "Initial Problem Decomposition",
##   "content": "The task involves counting the number of times the letter 'R' appears in the word 'strawberry'. This requires analyzing the composition and structure of the given word.",
##   "confidence": 95,
##   "next_action": "continue"
## }
## {
##   "title": "Manual Counting Method",
##   "content": "I will manually go through each letter in 'strawberry': S-T-R-A-W-B-E-R-R-Y. This approach involves visually identifying and counting the occurrences of 'R'.",
##   "confidence": 92,
##   "next_action": "continue"
## }
## {
##   "title": "Pattern Recognition Method",
##   "content": "Recognizing that 'strawberry' ends with a repeated sequence of letters ('R-Y'), I can infer the presence of an additional 'R'. This step leverages understanding of linguistic patterns.",
##   "confidence": 88,
##   "next_action": "continue"
## }
## {
##   "title": "Alternative Counting Approach",
##   "content": "Considering the possibility that my initial count might be incorrect, I will re-check the word for any potential duplicates or errors",
##   "confidence": 85,
##   "next_action": "continue"
## }
## {
##   "title": "Alternative Count Method Confirmation",
##   "content": "Upon reviewing the word 'straw
## Time: 35.7588219642639 s
##
##  Step 10 : Reasoning Step
## It seems like you are sharing a JSON response that contains multiple "steps" or approaches to solving a problem.
##
## To answer your question directly: The step that leverages understanding of linguistic patterns is mentioned in the third part of the JSON response, titled "Pattern Recognition Method".
##
## Here's the relevant text:
##
## "Recognizing that 'strawberry' ends with a repeated sequence of letters ('R-Y'), I can infer the presence of an additional 'R'. This step leverages understanding of linguistic patterns."
##
## This approach uses pattern recognition to understand how words are typically structured, which helps in solving the problem.
## Time: 19.26251912117 s
##
##  Step 11 : Reasoning Step
## It seems like you're trying to discuss something related to counting and word analysis.
##
## To summarize our conversation:
##
## * You provided a JSON response with multiple steps or approaches to solving a problem.
## * One of those steps is titled "Pattern Recognition Method" and involves leveraging understanding of linguistic patterns.
## * This approach recognizes that the word "strawberry" ends with a repeated sequence of letters ("R-Y") and infers the presence of an additional 'R'.
##
## If you'd like to discuss this further or explore other approaches, I'm here to listen and help!
## Time: 18.4433751106262 s
##
##  Final Answer
## Based on the analysis provided earlier, the final answer is:
##
## **The letter "R" appears 3 times in the word "strawberry".**
##
## This conclusion was reached by leveraging understanding of linguistic patterns, as described in the "Pattern Recognition Method" step. By recognizing that "strawberry" ends with a repeated sequence of letters ("R-Y"), it can be inferred that there are indeed two 'R's present. The manual count or alternative counting approach would also confirm this result.
##
## Therefore, the final answer is 3!
## Time: 19.0463080406189 s
## Total thinking time: 288.89 s

A: Therefore, the final answer is 3!


Note

This output is cherry-picked from llama3.1… because 70b was a lot to download for my internet connection.

Acknowledgments

Cursor is a great tool <3

Credits

License

This project is licensed under the MIT License

RStudio Addin

ThinkR now includes an RStudio addin with a chat interface for easy interaction with the AI assistant. To use the addin:

  1. Go to Addins > ThinkR Assistant in the RStudio menu.
  2. A chat interface will open in a new window.
  3. Enter your question or code in the text area and click "Submit".
  4. The AI's response will be displayed in the chat history.

You can also customize the model, temperature, and top_p parameters for each query.