Instructions to use TroyDoesAI/MermaidMistral with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use TroyDoesAI/MermaidMistral with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="TroyDoesAI/MermaidMistral") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("TroyDoesAI/MermaidMistral") model = AutoModelForCausalLM.from_pretrained("TroyDoesAI/MermaidMistral") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use TroyDoesAI/MermaidMistral with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "TroyDoesAI/MermaidMistral" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "TroyDoesAI/MermaidMistral", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/TroyDoesAI/MermaidMistral
- SGLang
How to use TroyDoesAI/MermaidMistral with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "TroyDoesAI/MermaidMistral" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "TroyDoesAI/MermaidMistral", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "TroyDoesAI/MermaidMistral" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "TroyDoesAI/MermaidMistral", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use TroyDoesAI/MermaidMistral with Docker Model Runner:
docker model run hf.co/TroyDoesAI/MermaidMistral
Commit ·
4069ae8
1
Parent(s): d6b12a7
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,63 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
### Introduction:
|
| 6 |
+
|
| 7 |
+
Introducing MermaidMistral, a powerful yet compact 7-billion-parameter language model adept at Python code understanding and crafting engaging story flow maps. Trained on a meticulously hand curated dataset of 478 diverse Python examples and hand crafted mermaid flow maps utilizing https://mermaid.live, this model goes beyond its size to deliver exceptional performance in code understanding and story visualization.
|
| 8 |
+
|
| 9 |
+
### Key Features:
|
| 10 |
+
MermaidMistral is not a "Chatty Kathy" and should only respond with a mermaid code block with a flow diagram in mermaid js syntax and nothing more.
|
| 11 |
+
|
| 12 |
+
**1. Code Understanding:**
|
| 13 |
+
- Grasps Python intricacies with finesse.
|
| 14 |
+
- Generates clear and accurate Mermaid Diagram Flow Charts.
|
| 15 |
+
- Ideal for developers seeking visual representations of their code's logic.
|
| 16 |
+
|
| 17 |
+
**2. Storytelling Capabilities:**
|
| 18 |
+
- Converts narrative inputs into captivating Mermaid Diagrams.
|
| 19 |
+
- Maps character interactions, plot developments, and narrative arcs effortlessly.
|
| 20 |
+
|
| 21 |
+
**3. Unmatched Performance:**
|
| 22 |
+
- Surpasses larger models, like GPT-4, in generating well-organized and detailed Mermaid Diagrams for story flows.
|
| 23 |
+
|
| 24 |
+
**4. Training Insights:**
|
| 25 |
+
- Trained on a 478 Python examples for just under three epochs on a single RTX 3090 following batch size equal to 1, known as stochastic gradient descent.
|
| 26 |
+
- Exhibited emergent properties in story-to-flow map translations.
|
| 27 |
+
- Adaptable and efficient in resource utilization
|
| 28 |
+
- Due to hardware constraints this fine tune has a token limit of 2048.
|
| 29 |
+
|
| 30 |
+
### Collaboration:
|
| 31 |
+
|
| 32 |
+
MermaidMistral is open to collaboration to further strengthen its capabilities. The dataset, formatted in Alpaca, provides a unique foundation for understanding Python intricacies. If you're interested in contributing or collaborating to enhance the model's performance, feel free to reach out to [troydoesai@gmail.com](mailto:troydoesai@gmail.com). Your expertise could play a pivotal role in refining MermaidMistral.
|
| 33 |
+
|
| 34 |
+
### Example Use Cases:
|
| 35 |
+
|
| 36 |
+
**1. Code Documentation:**
|
| 37 |
+
- Developers can use MermaidMistral to automatically generate visual flow charts from their Python code, aiding in documentation and code understanding.
|
| 38 |
+
|
| 39 |
+
**2. Storyboarding:**
|
| 40 |
+
- Storytellers and writers can input their narrative and receive visually appealing Mermaid Diagrams, offering a structured overview of character interactions and plot progression.
|
| 41 |
+
|
| 42 |
+
**3. Project Planning:**
|
| 43 |
+
- Project managers can leverage MermaidMistral to create visual project flow maps, facilitating effective communication and planning among team members.
|
| 44 |
+
|
| 45 |
+
**4. Learning Python:**
|
| 46 |
+
- Students and beginners can use MermaidMistral to visually understand Python code structures, enhancing their learning experience.
|
| 47 |
+
|
| 48 |
+
**5. Game Design:**
|
| 49 |
+
- Game developers can utilize MermaidMistral for visualizing game storylines, ensuring a coherent narrative structure and character development.
|
| 50 |
+
|
| 51 |
+
### Proof of Concept:
|
| 52 |
+
|
| 53 |
+
MermaidMistral proves that innovation thrives in compact packages, delivering exceptional performance across diverse applications. Its adaptability and efficiency showcase the potential for groundbreaking results even in resource-constrained environments.
|
| 54 |
+
|
| 55 |
+
### Example Story -> Flow
|
| 56 |
+
https://chat.openai.com/share/e3163857-981b-4968-b2db-98ad869c9259
|
| 57 |
+
|
| 58 |
+
### Insights on how to get best results
|
| 59 |
+
# For best results use full precision using one of the 3 different instruction types:
|
| 60 |
+
|
| 61 |
+
- "instruction": "Create the mermaid diagram for the following code:",
|
| 62 |
+
- "instruction": "Create the mermaid diagram for the following story:",
|
| 63 |
+
- "instruction": "Create the mermaid diagram for the following:",
|