Feat: Project files

adapters/README.md

@@ -0,0 +1,202 @@
+---
+base_model: microsoft/phi-2
+library_name: peft
+---
+
+# 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. -->
+
+
+
+- **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]
+### Framework versions
+
+- PEFT 0.14.0

adapters/adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "microsoft/phi-2",
+  "bias": "none",
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "dense",
+    "v_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapters/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8a057de8de7bf5b6239e976ad62100052736939a014dd69d06e3d67ee97743f
+size 41977360

adapters/added_tokens.json

@@ -0,0 +1,40 @@
+{
+  "\t\t": 50294,
+  "\t\t\t": 50293,
+  "\t\t\t\t": 50292,
+  "\t\t\t\t\t": 50291,
+  "\t\t\t\t\t\t": 50290,
+  "\t\t\t\t\t\t\t": 50289,
+  "\t\t\t\t\t\t\t\t": 50288,
+  "\t\t\t\t\t\t\t\t\t": 50287,
+  "  ": 50286,
+  "   ": 50285,
+  "    ": 50284,
+  "     ": 50283,
+  "      ": 50282,
+  "       ": 50281,
+  "        ": 50280,
+  "         ": 50279,
+  "          ": 50278,
+  "           ": 50277,
+  "            ": 50276,
+  "             ": 50275,
+  "              ": 50274,
+  "               ": 50273,
+  "                ": 50272,
+  "                 ": 50271,
+  "                  ": 50270,
+  "                   ": 50269,
+  "                    ": 50268,
+  "                     ": 50267,
+  "                      ": 50266,
+  "                       ": 50265,
+  "                        ": 50264,
+  "                         ": 50263,
+  "                          ": 50262,
+  "                           ": 50261,
+  "                            ": 50260,
+  "                             ": 50259,
+  "                              ": 50258,
+  "                               ": 50257
+}

adapters/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

adapters/tokenizer_config.json

@@ -0,0 +1,326 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50257": {
+      "content": "                               ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50258": {
+      "content": "                              ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50259": {
+      "content": "                             ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50260": {
+      "content": "                            ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50261": {
+      "content": "                           ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50262": {
+      "content": "                          ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50263": {
+      "content": "                         ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50264": {
+      "content": "                        ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50265": {
+      "content": "                       ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50266": {
+      "content": "                      ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50267": {
+      "content": "                     ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50268": {
+      "content": "                    ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50269": {
+      "content": "                   ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50270": {
+      "content": "                  ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50271": {
+      "content": "                 ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50272": {
+      "content": "                ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50273": {
+      "content": "               ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50274": {
+      "content": "              ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50275": {
+      "content": "             ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50276": {
+      "content": "            ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50277": {
+      "content": "           ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50278": {
+      "content": "          ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50279": {
+      "content": "         ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50280": {
+      "content": "        ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50281": {
+      "content": "       ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50282": {
+      "content": "      ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50283": {
+      "content": "     ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50284": {
+      "content": "    ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50285": {
+      "content": "   ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50286": {
+      "content": "  ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50287": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50288": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50289": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50290": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50291": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50292": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50293": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50294": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "model_max_length": 2048,
+  "pad_token": "<|endoftext|>",
+  "return_token_type_ids": false,
+  "tokenizer_class": "CodeGenTokenizer",
+  "unk_token": "<|endoftext|>"
+}

app.py

@@ -0,0 +1,234 @@
+#!/usr/bin/env python3
+"""
+Gradio application for inference with Phi-2 model using LoRA/QLoRA adapters.
+Pre-loads the model and provides a simple chat interface.
+"""
+
+import os
+import time
+import torch
+import gradio as gr
+from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
+from peft import PeftModel
+
+# Define constants
+DEFAULT_MODEL_PATH = "./adapters"  # Path to the trained adapters
+DEFAULT_BASE_MODEL = "microsoft/phi-2"  # Base model name
+DEFAULT_MAX_NEW_TOKENS = 512
+DEFAULT_TEMPERATURE = 0.7
+DEFAULT_TOP_P = 0.9
+DEFAULT_TOP_K = 50
+
+# Global variables to store the model and tokenizer
+model = None
+tokenizer = None
+
+def load_model(
+    model_path=DEFAULT_MODEL_PATH,
+    base_model=DEFAULT_BASE_MODEL,
+    use_qlora=True,
+    device="cuda"
+):
+    """
+    Load the base model and adapter weights.
+    """
+    global model, tokenizer
+    
+    print(f"Loading tokenizer from {base_model}...")
+    tokenizer = AutoTokenizer.from_pretrained(base_model, trust_remote_code=True)
+    if tokenizer.pad_token is None:
+        tokenizer.pad_token = tokenizer.eos_token
+    
+    # Configure model loading parameters
+    model_kwargs = {"trust_remote_code": True}
+    
+    # Set up quantization for QLoRA if enabled
+    if use_qlora:
+        print("Using 4-bit quantization (QLoRA)")
+        compute_dtype = torch.float16
+        if torch.cuda.is_bf16_supported():
+            compute_dtype = torch.bfloat16
+            
+        quantization_config = BitsAndBytesConfig(
+            load_in_4bit=True,
+            bnb_4bit_quant_type="nf4",
+            bnb_4bit_compute_dtype=compute_dtype,
+            bnb_4bit_use_double_quant=True
+        )
+        model_kwargs["quantization_config"] = quantization_config
+    else:
+        model_kwargs["torch_dtype"] = torch.float16 if torch.cuda.is_available() else torch.float32
+    
+    # Check if adapter path exists
+    if not os.path.exists(model_path):
+        print(f"Warning: Model path '{model_path}' does not exist. Using base model only.")
+    
+    # Load base model
+    print(f"Loading base model {base_model}...")
+    base_model = AutoModelForCausalLM.from_pretrained(
+        base_model, 
+        **model_kwargs
+    )
+    
+    # Load adapter weights if available
+    if os.path.exists(model_path) and os.path.exists(os.path.join(model_path, "adapter_config.json")):
+        print(f"Loading {'QLoRA' if use_qlora else 'LoRA'} adapters from {model_path}...")
+        model = PeftModel.from_pretrained(base_model, model_path)
+        
+        # Special handling for QLoRA - move norm layers to float32 for stability
+        # and ensure model and adapter layers have consistent dtypes
+        if use_qlora:
+            print("Harmonizing model layer dtypes...")
+            working_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
+            
+            # First make sure important parts are in float16/32
+            for name, module in model.named_modules():
+                if any(x in name for x in ["lm_head", "embed_tokens"]):
+                    module.to(working_dtype)
+                elif "norm" in name:
+                    module.to(torch.float32)  # Norms should be in fp32 for stability
+    else:
+        model = base_model
+        print("Using base model without adapters")
+    
+    # Move model to device
+    device = torch.device(device if torch.cuda.is_available() else "cpu")
+    model = model.to(device)
+    model.eval()
+    
+    print(f"Model loaded successfully and moved to {device}!")
+    return model, tokenizer
+
+
+def generate_response(prompt, chat_history):
+    """
+    Generate text response from the model.
+    """
+    global model, tokenizer
+    
+    if model is None or tokenizer is None:
+        return chat_history + [(prompt, "Model not loaded yet. Please wait a moment.")]
+    
+    # Format prompt for Phi-2
+    formatted_prompt = f"Instruct: {prompt}\nOutput:"
+    
+    # Tokenize input prompt
+    device = next(model.parameters()).device
+    input_ids = tokenizer.encode(formatted_prompt, return_tensors="pt").to(device)
+    attention_mask = torch.ones_like(input_ids).to(device)
+    
+    # Generate text with robust error handling
+    try:
+        with torch.no_grad():
+            # Explicit type casting
+            input_ids = input_ids.to(torch.long)  # IDs should always be long
+            attention_mask = attention_mask.to(torch.float16 if torch.cuda.is_available() else torch.float32)
+            
+            # First attempt with simple parameters
+            generated_ids = model.generate(
+                input_ids=input_ids,
+                max_new_tokens=DEFAULT_MAX_NEW_TOKENS,
+                do_sample=True,
+                temperature=DEFAULT_TEMPERATURE,
+                top_p=DEFAULT_TOP_P,
+                top_k=DEFAULT_TOP_K,
+            )
+    except Exception as e:
+        print(f"Generation error: {str(e)}")
+        try:
+            # Fallback: Try with model in eval with forced dtype
+            print("Attempting fallback generation...")
+            with torch.autocast(device_type='cuda' if torch.cuda.is_available() else 'cpu', 
+                              dtype=torch.float16 if torch.cuda.is_available() else torch.float32):
+                generated_ids = model.generate(
+                    input_ids=input_ids,
+                    max_new_tokens=DEFAULT_MAX_NEW_TOKENS,
+                    do_sample=False,  # Use greedy decoding for more stability
+                )
+        except Exception as e2:
+            return chat_history + [(prompt, f"Error generating response: {str(e2)}")]
+    
+    # Decode the generated text
+    generated_text = tokenizer.decode(generated_ids[0], skip_special_tokens=True)
+    
+    # Extract just the output part
+    output = generated_text.split("Output:")[1].strip() if "Output:" in generated_text else generated_text
+    
+    # Update chat history
+    return chat_history + [(prompt, output)]
+
+
+# Example prompts to demonstrate model capabilities
+examples = [
+    ["Explain the concept of quantum computing in simple terms."],
+    ["Write a short story about a robot that learns to paint."],
+    ["What are some ethical considerations when developing AI systems?"],
+    ["How can I improve my productivity while working from home?"],
+    ["Create a meal plan for a vegetarian diet that provides sufficient protein."]
+]
+
+
+# Initialize the model at startup
+print("Pre-loading the model...")
+try:
+    model, tokenizer = load_model()
+except Exception as e:
+    print(f"Error loading model: {str(e)}")
+    print("The app will still start, but you may need to check your model path.")
+
+# Create the Gradio interface
+with gr.Blocks(title="Supervised Fine Tuned (SFT) Phi-2 with QLoRA Adapters") as demo:
+    gr.Markdown("# Supervised Fine Tuned (SFT) Phi-2 with QLoRA Adapters")
+    gr.Markdown("- Base model (foundation model) Phi-2\n"
+                 "- Supervised Fine Tuned (SFT) method is used to fine-tune the model on [OpenAssistant dataset](https://huggingface.co/datasets/OpenAssistant/oasst1?row=0)\n"
+                 "- QLoRA Adapters are used to reduce the number of parameters in the model\n"
+                 "- This gives the model an ability to answer questions rather than just generating text\n"
+                 "- Chat with SFT Phi-2 model with QLoRA Adapters")
+    
+    chatbot = gr.Chatbot(height=500)
+    
+    with gr.Row():
+        msg = gr.Textbox(
+            label="Type your message here", 
+            placeholder="Ask me anything...",
+            show_label=False,
+            scale=9
+        )
+        send_btn = gr.Button("Send", scale=1)
+    
+    clear = gr.Button("Clear Chat")
+    
+    # Add examples section
+    gr.Markdown("### Example Capabilities")
+    gr.Examples(
+        examples=examples,
+        inputs=msg,
+        outputs=chatbot,
+        fn=generate_response,
+        cache_examples=False,
+        examples_per_page=5
+    )
+    
+    # Set up event handlers
+    send_btn.click(generate_response, [msg, chatbot], [chatbot]).then(
+        lambda: "", None, msg  # Clear the input box after sending
+    )
+    
+    msg.submit(generate_response, [msg, chatbot], [chatbot]).then(
+        lambda: "", None, msg  # Clear the input box after sending
+    )
+    
+    clear.click(lambda: [], None, chatbot)
+
+# Launch the app
+if __name__ == "__main__":
+    # Check GPU status
+    if torch.cuda.is_available():
+        print(f"CUDA available: {torch.cuda.get_device_name(0)}")
+        print(f"Memory allocated: {torch.cuda.memory_allocated() / 1024**3:.2f} GB")
+        print(f"Memory reserved: {torch.cuda.memory_reserved() / 1024**3:.2f} GB")
+    else:
+        print("CUDA not available, using CPU. This will be very slow for inference.")
+    
+    # Launch the Gradio app
+    demo.launch(share=True) 

requirements.txt

@@ -0,0 +1,31 @@
+# Core dependencies
+torch>=2.0.0
+transformers>=4.37.0
+datasets>=2.12.0
+accelerate>=0.20.0
+peft>=0.5.0
+
+# Quantization and optimization
+bitsandbytes>=0.40.0
+optimum>=1.12.0
+safetensors>=0.3.1
+
+# Training frameworks
+pytorch-lightning>=2.0.0
+deepspeed>=0.10.0
+
+# Monitoring and logging
+tensorboard>=2.12.0
+wandb>=0.15.0
+matplotlib>=3.7.0
+evaluate>=0.4.0
+
+# Data processing
+nltk>=3.8.0
+pandas>=2.0.0
+scipy>=1.10.0
+tqdm>=4.65.0
+
+# Optional: For better text processing
+sentencepiece>=0.1.99
+tokenizers>=0.13.3