daekeun-ml
/

Phi-3-medium-4k-instruct-ko-poc-v0.1

@@ -1,200 +1,112 @@
 ---
 library_name: transformers
 tags:
 - unsloth
 ---
-# Model Card for Model ID
-<!-- Provide a quick summary of what the model is/does. -->
 ## Model Details
-### Model Description
-<!-- Provide a longer summary of what this model is. -->
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-### Model Sources [optional]
-<!-- Provide the basic links for the model. -->
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-## Uses
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-### Direct Use
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-[More Information Needed]
-### Downstream Use [optional]
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-[More Information Needed]
-### Out-of-Scope Use
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-[More Information Needed]
-## Bias, Risks, and Limitations
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-[More Information Needed]
-### Recommendations
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
 ## How to Get Started with the Model
-Use the code below to get started with the model.
-[More Information Needed]
-## Training Details
-### Training Data
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-[More Information Needed]
-### Training Procedure
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-#### Preprocessing [optional]
-[More Information Needed]
-#### Training Hyperparameters
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-#### Speeds, Sizes, Times [optional]
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-[More Information Needed]
-## Evaluation
-<!-- This section describes the evaluation protocols and provides the results. -->
-### Testing Data, Factors & Metrics
-#### Testing Data
-<!-- This should link to a Dataset Card if possible. -->
-[More Information Needed]
-#### Factors
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-[More Information Needed]
-#### Metrics
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-[More Information Needed]
-### Results
-[More Information Needed]
-#### Summary
-## Model Examination [optional]
-<!-- Relevant interpretability work for the model goes here -->
-[More Information Needed]
-## Environmental Impact
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-## Technical Specifications [optional]
-### Model Architecture and Objective
-[More Information Needed]
-### Compute Infrastructure
-[More Information Needed]
-#### Hardware
-[More Information Needed]
-#### Software
-[More Information Needed]
-## Citation [optional]
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-**BibTeX:**
-[More Information Needed]
-**APA:**
-[More Information Needed]
-## Glossary [optional]
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-[More Information Needed]
-## More Information [optional]
-[More Information Needed]
-## Model Card Authors [optional]
-[More Information Needed]
-## Model Card Contact
-[More Information Needed]

 ---
 library_name: transformers
+license: apache-2.0
+datasets:
+- nampdn-ai/tiny-codes
+- nlpai-lab/openassistant-guanaco-ko
+- philschmid/guanaco-sharegpt-style
+language:
+- ko
+- en
+inference: false
 tags:
 - unsloth
+- phi-3
+pipeline_tag: text-generation
 ---
+# Phi-3-medium-4k-instruct-ko-poc-v0.1
 ## Model Details
+This model is trained using unsloth toolkit based on Microsoft's phi-3 model with some Korean instruction data added to enhance its Korean generation performance
+Since my role is not as a working developer, but as ML Technical Specialist helping customers with quick PoCs/prototypes, and I was limited by Azure GPU resources available, I only trained with 40,000 samples on a single A100 GPU () for PoC purposes. Because I have not done any tokenizer extensions, you need a lot more tokens than English for text generation.
+### Dataset
+The dataset used for training is as follows. To prevent catastrophic forgetting, I included non-Korean corpus as training data. Note that we did not use all of the data, but only sampled some of it. Korean textbooks were converted to Q&A format. The Guanaco dataset has been reformatted to fit the multiturn format like <|user|>\n{Q1}<|end|>\n<|assistant|>\n{A1}<|end|>\n<|user|>\n{Q2}<|end|>\n<|assistant|>\n{A2}<|end|>.
+- Korean textbooks (https://huggingface.co/datasets/nampdn-ai/tiny-codes)
+- Korean translation of Guanaco (https://huggingface.co/datasets/nlpai-lab/openassistant-guanaco-ko)
+- Guanaco Sharegpt style (https://huggingface.co/datasets/philschmid/guanaco-sharegpt-style)
 ## How to Get Started with the Model
+```python
+### Load model
+import torch
+from unsloth import FastLanguageModel
+from unsloth.chat_templates import get_chat_template
+from transformers import TextStreamer
+max_seq_length = 2048 # Choose any! We auto support RoPE Scaling internally!
+dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
+load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.
+model_path = "daekeun-ml/Phi-3-medium-4k-instruct-ko-poc-v0.1"
+model, tokenizer = FastLanguageModel.from_pretrained(
+    model_name = model_tar_dir, # Choose ANY! eg teknium/OpenHermes-2.5-Mistral-7B
+    max_seq_length = max_seq_length,
+    dtype = dtype,
+    load_in_4bit = load_in_4bit,
+    # token = "hf_...", # use one if using gated models like meta-llama/Llama-2-7b-hf
+)
+tokenizer = get_chat_template(
+    tokenizer,
+    chat_template = "phi-3", # Supports zephyr, chatml, mistral, llama, alpaca, vicuna, vicuna_old, unsloth
+    mapping = {"role" : "from", "content" : "value", "user" : "human", "assistant" : "gpt"}, # ShareGPT style
+)
+params = {
+    "max_new_tokens": 256,
+    "use_cache": True,
+    "temperature": 0.05,
+    "do_sample": True
+}
+### Inference
+FastLanguageModel.for_inference(model) # Enable native 2x faster inference
+messages = [
+    {"from": "human", "value": "Continue the fibonnaci sequence in Korean: 1, 1, 2, 3, 5, 8,"},
+    {"from": "assistant", "value": "피보나치 수열의 다음 숫자는 13, 21, 34, 55, 89 등입니다. 각 숫자는 앞의 두 숫자의 합입니다."},
+    {"from": "human", "value": "Compute 2x+3=12 in Korean"},
+]
+inputs = tokenizer.apply_chat_template(
+    messages,
+    tokenize = True,
+    add_generation_prompt = True, # Must add for generation
+    return_tensors = "pt",
+).to("cuda")
+text_streamer = TextStreamer(tokenizer)
+_ = model.generate(input_ids = inputs, streamer = text_streamer, **params)
+messages = [
+    {"from": "human", "value": "What is Machine Learning in Korean?"},
+    {"from": "assistant", "value": "인공지능의 한 분야로 방대한 데이터를 분석해 향후 패턴을 예측하는 기법입니다."},
+    {"from": "human", "value": "What is Deep Learning in Korean?"},
+]
+inputs = tokenizer.apply_chat_template(
+    messages,
+    tokenize = True,
+    add_generation_prompt = True, # Must add for generation
+    return_tensors = "pt",
+).to("cuda")
+from transformers import TextStreamer
+text_streamer = TextStreamer(tokenizer)
+_ = model.generate(input_ids = inputs, streamer = text_streamer, **params)
+```
+### References
+- Base model: [microsoft/phi-2](https://huggingface.co/microsoft/phi-2)
+## Notes
+### License
+apache 2.0; The license of phi-3 is MIT, but I considered the licensing of the dataset and library used for training.
+### Caution
+This model was created as a personal experiment, unrelated to the organization I work for. The model may not operate correctly because separate verification was not performed. Please be careful unless it is for personal experimentation or PoC (Proof of Concept)!