Update README.md

README.md

@@ -19,36 +19,68 @@ This is the model card of a 🤗 transformers model that has been pushed on the
 - **Developed by:** HuyRemy
 - **Funded by [optional]:** HuyRemy
 - **Shared by [optional]:** HuyRemy
-- **Model type:** Alpaca
+- **Model type:** Mistral
 - **License:** [email protected]
 
 
 ### Model Sources [optional]
 
-- **USE T4 GPU
+- **
 - **Demo [optional]:** https://matilda.vn
 
 ## Uses
 
+USE T4 GPU
+```Python
 !pip install "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"
-
+!pip install --no-deps xformers trl peft accelerate bitsandbytes
+```
 ### Direct Use
 ``` Python
 from unsloth import FastLanguageModel
+import torch
+max_seq_length = 2048
+dtype = None 
+load_in_4bit = True 
+alpaca_prompt = """Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
+
+### Instruction:
+{}
+
+### Input:
+{}
+
+### Response:
+{}"""
+
+
+EOS_TOKEN = tokenizer.eos_token # Must add EOS_TOKEN
+def formatting_prompts_func(examples):
+    instructions = examples["instruction"]
+    inputs       = examples["input"]
+    outputs      = examples["output"]
+    texts = []
+    for instruction, input, output in zip(instructions, inputs, outputs):
+        # Must add EOS_TOKEN, otherwise your generation will go on forever!
+        text = alpaca_prompt.format(instruction, input, output) + EOS_TOKEN
+        texts.append(text)
+    return { "text" : texts, }
+pass
+
 model, tokenizer = FastLanguageModel.from_pretrained(
-    model_name = "huyremy/aichat", # YOUR MODEL YOU USED FOR TRAINING
+    model_name = "huyremy/aichat", 
     max_seq_length = max_seq_length,
     dtype = dtype,
     load_in_4bit = load_in_4bit,
 )
-FastLanguageModel.for_inference(model) # Enable native 2x faster inference
+FastLanguageModel.for_inference(model) 
 
 inputs = tokenizer(
 [
     alpaca_prompt.format(
-        "who is Nguyễn Phú Trọng?", # instruction
-        "", # input
-        "", # output - leave this blank for generation!
+        "who is Nguyễn Phú Trọng?", 
+        "", 
+        "", 
     ),
 ], return_tensors = "pt").to("cuda")