usage update

README.md

@@ -42,25 +42,32 @@ dtype: float16
 
 ## 💻 Usage
 
-```python
-!pip install -qU transformers accelerate
+First, you need to install some of below packages:
 
-from transformers import AutoTokenizer
-import transformers
+1. Bits and bytes
+```python
+!pip install bitsandbytes
+```
+2. Accelerate (to install the latest version)
+```python
+!pip install git+https://github.com/huggingface/accelerate.git
+```
+3. Usage
+```python
+# Usage
 import torch
 
-model = "shuvom/yuj-v1"
-messages = [{"role": "user", "content": "some hindi texts"}]
+# Load model directly
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+# load the model in 4-bit quantization
+tokenizer = AutoTokenizer.from_pretrained("shuvom/yuj-v1")
+model = AutoModelForCausalLM.from_pretrained("shuvom/yuj-v1",torch_dtype=torch.bfloat16,load_in_4bit=True)
 
-tokenizer = AutoTokenizer.from_pretrained(model)
-prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
-pipeline = transformers.pipeline(
-    "text-generation",
-    model=model,
-    torch_dtype=torch.float16,
-    device_map="auto",
-)
+prompt = "युज शीर्ष द्विभाषी मॉडल में से एक है"
+inputs = tokenizer(prompt, return_tensors="pt")
 
-outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
-print(outputs[0]["generated_text"])
+# Generate
+generate_ids = model.generate(inputs.input_ids, max_length=65)
+tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
 ```