Update README.md

README.md

@@ -31,9 +31,75 @@ inference: true
 > باستخدام مكتبة `transformers`:
 
 ```python
-from transformers import AutoTokenizer, AutoModelForCausalLM, TextStreamer
+!pip install --upgrade bitsandbytes
+!pip install -q datasets
+!pip install -q trl
+!pip install git+https://github.com/huggingface/peft.git
+!pip install -q -U accelerate
 
-model_id = "<username>/<repo-name>"  # هذا المستودع
-tok = AutoTokenizer.from_pretrained(model_id, use_fast=True)
-model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="auto", device_map="auto")
 
+from huggingface_hub import login
+login()
+
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
+from peft import LoraConfig, get_peft_model, prepare_model_for_kbit_training, PeftModel, PeftConfig
+from datasets import load_dataset
+from transformers import TrainingArguments, pipeline
+from trl import SFTTrainer
+
+bnb_cfg = BitsAndBytesConfig(
+    load_in_4bit=True,
+    bnb_4bit_quant_type="nf4",
+    bnb_4bit_use_double_quant=True,
+    bnb_4bit_compute_dtype="bfloat16",
+)
+
+# -*- coding: utf-8 -*-
+
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM
+model_path = "3okasha/jais-finetuned-v1"
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+tokenizer = AutoTokenizer.from_pretrained(model_path)
+model = AutoModelForCausalLM.from_pretrained(
+    model_path, 
+    quantization_config=bnb_cfg,
+    device_map="auto", 
+    trust_remote_code=True
+    )
+
+def user_prompt(human_prompt):
+    prompt_template=f"### HUMAN:\n{human_prompt}\n\n### RESPONSE:\n"
+    return prompt_template
+
+
+model.config.use_cache = False
+if hasattr(model, "generation_config"): model.generation_config.use_cache = False
+
+def get_response(text,tokenizer=tokenizer,model=model):
+    input_ids = tokenizer(text, return_tensors="pt").input_ids
+    inputs = input_ids.to(device)
+    input_len = inputs.shape[-1]
+    generate_ids = model.generate(
+        inputs,
+        top_p=0.9,
+        temperature=0.3,
+        max_length=50-input_len,
+        min_length=input_len + 4,
+        repetition_penalty=1.2,
+        do_sample=True,
+    )
+    response = tokenizer.batch_decode(
+        generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=True
+    )[0]
+    return response
+
+
+text= user_prompt("كيف الحال")
+print(get_response(text))
+
+
+```