Update README.md

README.md

@@ -1,199 +1,215 @@
 ---
-library_name: transformers
-tags: []
+datasets:
+- homebrewltd/instruction-speech-whispervq-v2
+language:
+- en
+license: apache-2.0
+tags:
+- sound language model
 ---
 
-# 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
+We have developed and released the family [Ichigo-llama3s](https://huggingface.co/collections/homebrew-research/llama3-s-669df2139f0576abc6eb7405). This family is natively understanding audio and text input.
 
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+This model focused on fine-tuning the model to improve user interaction from [homebrewltd/Ichigo-llama3.1-s-instruct-v0.3-phase-2](https://huggingface.co/homebrewltd/Ichigo-llama3.1-s-instruct-v0.3-phase-2), particularly in handling inaudible inputs and multi-turn conversations.
 
-### Direct Use
+**Model developers** Homebrew Research.
 
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+**Input** Text and sound.
 
-[More Information Needed]
+**Output** Text.
 
-### Downstream Use [optional]
+**Model Architecture** Llama-3.
 
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+**Language(s):** English.
 
-[More Information Needed]
+## Intended Use
 
-### Out-of-Scope Use
+**Intended Use Cases** This family is primarily intended for research applications. This version aims to further improve the LLM on sound understanding capabilities.
 
-<!-- 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.
+**Out-of-scope** The use of llama3-s in any manner that violates applicable laws or regulations is strictly prohibited.
 
 ## How to Get Started with the Model
 
-Use the code below to get started with the model.
-
-[More Information Needed]
-
-## Training Details
-
-### Training Data
+Try this model using [Google Colab Notebook](https://colab.research.google.com/drive/18IiwN0AzBZaox5o0iidXqWD1xKq11XbZ?usp=sharing).
+
+First, we need to convert the audio file to sound tokens
+
+```python
+device = "cuda" if torch.cuda.is_available() else "cpu"
+if not os.path.exists("whisper-vq-stoks-medium-en+pl-fixed.model"):
+    hf_hub_download(
+        repo_id="jan-hq/WhisperVQ",
+        filename="whisper-vq-stoks-medium-en+pl-fixed.model",
+        local_dir=".",
+    )
+vq_model = RQBottleneckTransformer.load_model(
+        "whisper-vq-stoks-medium-en+pl-fixed.model"
+    ).to(device)
+vq_model.ensure_whisper(device)
+def audio_to_sound_tokens(audio_path, target_bandwidth=1.5, device=device):
+
+    wav, sr = torchaudio.load(audio_path)
+    if sr != 16000:
+        wav = torchaudio.functional.resample(wav, sr, 16000)
+    with torch.no_grad():
+        codes = vq_model.encode_audio(wav.to(device))
+        codes = codes[0].cpu().tolist()
+
+    result = ''.join(f'<|sound_{num:04d}|>' for num in codes)
+    return f'<|sound_start|>{result}<|sound_end|>'
+```
+
+Then, we can inference the model the same as any other LLM.
+
+```python
+def setup_pipeline(model_path, use_4bit=False, use_8bit=False):
+    tokenizer = AutoTokenizer.from_pretrained(model_path)
+
+    model_kwargs = {"device_map": "auto"}
+
+    if use_4bit:
+        model_kwargs["quantization_config"] = BitsAndBytesConfig(
+            load_in_4bit=True,
+            bnb_4bit_compute_dtype=torch.bfloat16,
+            bnb_4bit_use_double_quant=True,
+            bnb_4bit_quant_type="nf4",
+        )
+    elif use_8bit:
+        model_kwargs["quantization_config"] = BitsAndBytesConfig(
+            load_in_8bit=True,
+            bnb_8bit_compute_dtype=torch.bfloat16,
+            bnb_8bit_use_double_quant=True,
+        )
+    else:
+        model_kwargs["torch_dtype"] = torch.bfloat16
 
-<!-- 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. -->
+    model = AutoModelForCausalLM.from_pretrained(model_path, **model_kwargs)
 
-[More Information Needed]
+    return pipeline("text-generation", model=model, tokenizer=tokenizer)
 
-### Training Procedure
+def generate_text(pipe, messages, max_new_tokens=64, temperature=0.0, do_sample=False):
+    generation_args = {
+        "max_new_tokens": max_new_tokens,
+        "return_full_text": False,
+        "temperature": temperature,
+        "do_sample": do_sample,
+    }
 
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+    output = pipe(messages, **generation_args)
+    return output[0]['generated_text']
 
-#### Preprocessing [optional]
+# Usage
+llm_path = "homebrewltd/llama3.1-s-instruct-v0.2"
+pipe = setup_pipeline(llm_path, use_8bit=True)
+```
 
-[More Information Needed]
+## Training process
+**Training Metrics Image**: Below is a snapshot of the training loss curve visualized.
 
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/65713d70f56f9538679e5a56/7TWPqLdDLDlfzeRXP9m36.png)
 
-#### Training Hyperparameters
+**[MMLU](https://huggingface.co/datasets/cais/mmlu)**:
 
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+| Model | MMLU Score |
+| --- | --- |
+| llama3.5-instruct-8b | 69.40 |
+| ichigo-llama3.1-s-v0.3: phase 3 | **63.79** |
+| ichigo-llama3.1-s-v0.3: phase 2 | 63.08 |
+| ichigo-llama3.1-s-base-v0.3 | 42.11 |
+| llama3.5-instruct-v0.2 | 50.27 |
 
-#### Speeds, Sizes, Times [optional]
+**[AudioBench](https://arxiv.org/abs/2406.16020) Eval**:
 
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+| Model Bench | [Open-hermes Instruction Audio](https://huggingface.co/datasets/AudioLLMs/openhermes_instruction_test) (GPT-4-O judge 0:5) | [Alpaca Instruction Audio](https://huggingface.co/datasets/AudioLLMs/alpaca_audio_test) (GPT-4-O judge 0:5) |
+| --- | --- | --- |
+| [Llama3.1-s-v2](https://huggingface.co/homebrewltd/llama3-s-instruct-v0.2) | 3.45 | 3.53 |
+| [Ichigo-llama3.1-s v0.3-phase2 -cp7000](https://huggingface.co/homebrewltd/Ichigo-llama3.1-s-instruct-v0.3-phase-2) | 3.42 | 3.62 |
+| [Ichigo-llama3.1-s v0.3-phase2-cplast](https://huggingface.co/jan-hq/llama3-s-instruct-v0.3-checkpoint-last) | 3.31 | 3.6 |
+| [Ichigo-llama3.1-s v0.3-phase3](https://huggingface.co/homebrewltd/Ichigo-llama3.1-s-instruct-v0.3-phase-3) | **3.64** | **3.68** |
+| [Qwen2-audio-7B](https://huggingface.co/Qwen/Qwen2-Audio-7B) | 2.63 | 2.24 |
 
-[More Information Needed]
+### Hardware
 
-## Evaluation
+**GPU Configuration**: Cluster of 8x NVIDIA H100-SXM-80GB.
+**GPU Usage**:
+  - **Continual Training**: 3 hours.
 
-<!-- This section describes the evaluation protocols and provides the results. -->
+### Training Arguments
 
-### Testing Data, Factors & Metrics
+We utilize [torchtune](https://github.com/pytorch/torchtune) library for the latest FSDP2 training code implementation. 
 
-#### Testing Data
+| **Parameter** | **Continual Training** |
+| --- | --- |
+| **Epoch** | 1 |
+| **Global batch size** | 256 |
+| **Learning Rate** | 1.5e-5 |
+| **Learning Scheduler** | LambdaLR with warmup |
+| **Optimizer** | [AdamW Fused](https://pytorch.org/docs/stable/generated/torch.optim.AdamW.html) |
+| **Warmup Steps** | 8 |
+| **Weight Decay** | 0.005 |
+| **Max length** | 4096 |
+| **Precision** | bf16 |
 
-<!-- This should link to a Dataset Card if possible. -->
 
-[More Information Needed]
+## Examples
 
-#### Factors
+1. Good example:
 
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+<details>
+<summary>Click to toggle Example 1</summary>
 
-[More Information Needed]
+```
 
-#### Metrics
+```
+</details>
 
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+<details>
+<summary>Click to toggle Example 2</summary>
 
-[More Information Needed]
+```
 
-### Results
+```
+</details>
 
-[More Information Needed]
 
-#### Summary
+2. Misunderstanding example:
 
+<details>
+<summary>Click to toggle Example 3</summary>
+  
+```
 
+```
+</details>
 
-## Model Examination [optional]
+3. Off-tracked example:
 
-<!-- Relevant interpretability work for the model goes here -->
+<details>
+<summary>Click to toggle Example 4</summary>
 
-[More Information Needed]
+```
 
-## Environmental Impact
+```
+</details>
 
-<!-- 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. -->
+## Citation Information
 
 **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]
+```
+@article{Llama3-S: Sound Instruction Language Model 2024,
+  title={Llama3-S},
+  author={Homebrew Research},
+  year=2024,
+  month=August},
+  url={https://huggingface.co/homebrewltd/llama3.1-s-2024-08-20}
+```
 
-[More Information Needed]
+## Acknowledgement
 
-## Model Card Contact
+- **[WhisperSpeech](https://github.com/collabora/WhisperSpeech)**
 
-[More Information Needed]
+- **[Meta-Llama-3.1-8B-Instruct ](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct)**