Update README.md

README.md

@@ -4,6 +4,7 @@ language:
 - hi
 license: apache-2.0
 tags:
+- text-generation
 - transformers
 - unsloth
 - gemma
@@ -13,6 +14,7 @@ datasets:
 - yahma/alpaca-cleaned
 - ravithejads/samvaad-hi-filtered
 - HydraIndicLM/hindi_alpaca_dolly_67k
+pipeline_tag: text-generation
 ---
 
 # Gemma-2B-Hinglish-LORA-v1.0 model
@@ -20,7 +22,55 @@ datasets:
 - **Developed by:** [Kiran Kunapuli](https://www.linkedin.com/in/kirankunapuli/)
 - **License:** apache-2.0
 - **Finetuned from model :** unsloth/gemma-2b-bnb-4bit
-- - **Model config:**
+- **Model usage:** Use the below code in Python
+  ```python
+    import torch
+    from transformers import AutoTokenizer, AutoModelForCausalLM
+    
+    tokenizer = AutoTokenizer.from_pretrained("kirankunapuli/Gemma-2B-Hinglish-LORA-v1.0")
+    model = AutoModelForCausalLM.from_pretrained("kirankunapuli/Gemma-2B-Hinglish-LORA-v1.0")
+
+    device = "cuda:0" if torch.cuda.is_available() else "cpu"
+    model = model.to(device)
+
+    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:
+    {}"""
+
+    # Example 1
+    inputs = tokenizer(
+    [
+        alpaca_prompt.format(
+            "ऐतिहासिक स्मारक India Gate कहाँ स्थित है?", # instruction
+            "", # input
+            "", # output - leave this blank for generation!
+        )
+    ], return_tensors = "pt").to(device)
+    
+    outputs = model.generate(**inputs, max_new_tokens = 64, use_cache = True)
+    print(tokenizer.batch_decode(outputs))
+    
+    # Example 2
+    inputs = tokenizer(
+    [
+        alpaca_prompt.format(
+            "ऐतिहासिक स्मारक इंडिया गेट कहाँ स्थित है? मुझे अंग्रेजी में बताओ", # instruction
+            "", # input
+            "", # output - leave this blank for generation!
+        )
+    ], return_tensors = "pt").to(device)
+    
+    outputs = model.generate(**inputs, max_new_tokens = 64, use_cache = True)
+    print(tokenizer.batch_decode(outputs))
+  ```
+- **Model config:**
   ```python
     model = FastLanguageModel.get_peft_model(
     model,