Update README.md

README.md

@@ -1,21 +1,9 @@
----
-language: en
-tags:
-- text-generation
-- YouTube-scripts
-- fine-tuned
-- causal-lm
-datasets:
-- custom
-license: mit
-model_name: Gemma 2 Scripter
----
-
 # Gemma 2 Scripter
 
 **Gemma 2 Scripter** is a fine-tuned version of the Gemma 2 2B instruct model designed for generating high-quality YouTube scripts based on provided keywords. It is optimized for text generation tasks, delivering coherent and contextually relevant outputs.
 
 ## Model Details
+
 - **Model Name**: `Sidharthan/gemma2_scripter`
 - **Architecture**: Causal Language Model
 - **Base Model**: Gemma 2 2B
@@ -24,15 +12,16 @@ model_name: Gemma 2 Scripter
 ## How to Use
 
 ### Installation
+
 Ensure you have the following dependencies installed:
+
 ```bash
 pip install torch transformers peft
 ```
 
 ### Code Sample
 
-python```
-
+```python
 from transformers import AutoTokenizer
 from peft import AutoPeftModelForCausalLM
 import torch
@@ -41,6 +30,7 @@ import torch
 model_name = "Sidharthan/gemma2_scripter"
 tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
 device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
+
 model = AutoPeftModelForCausalLM.from_pretrained(
     model_name,
     device_map=None,
@@ -51,9 +41,10 @@ model = AutoPeftModelForCausalLM.from_pretrained(
 
 # Generate a script
 def generate_script(prompt):
-    formatted_prompt = f"<bos><start_of_turn>keywords\n{prompt}<end_of_turn>\n<start_of_turn>script\n"
+    formatted_prompt = f"keywords\n{prompt}\nscript\n"
     inputs = tokenizer(formatted_prompt, return_tensors="pt")
     inputs = {key: value.to(device) for key, value in inputs.items()}
+    
     outputs = model.generate(
         **inputs,
         max_length=1024,
@@ -65,6 +56,7 @@ def generate_script(prompt):
         pad_token_id=tokenizer.pad_token_id,
         eos_token_id=tokenizer.eos_token_id
     )
+    
     return tokenizer.decode(outputs[0], skip_special_tokens=True).strip()
 
 # Example usage
@@ -77,23 +69,17 @@ print(f"Generated Script:\n{response}")
 
 The model expects prompts in the following format:
 
-bash```
-
-<bos><start_of_turn>keywords
-<your_keywords_here><end_of_turn>
-<start_of_turn>script
-
+```
+keywords
+<your keywords here>
+script
 ```
 
 Example:
-
-bash```
-
-<bos><start_of_turn>keywords
-crosshatch waffle texture, dark chocolate, four bar crispy wafers, kat, milk chocolate<end_of_turn>
-<start_of_turn>script
-
-
+```
+keywords
+crosshatch waffle texture, dark chocolate, four bar crispy wafers, kat, milk chocolate
+script
 ```
 
 ### Output
@@ -105,11 +91,10 @@ The output is a YouTube script generated based on the keywords provided.
 - CPU: Slower inference due to computational constraints.
 - GPU: Optimized for faster inference with FP16 support.
 
-
 ### Applications
 
-- Generating structured scripts for video content.
-- Keyword-based text generation for creative tasks.
+- Generating structured scripts for video content
+- Keyword-based text generation for creative tasks
 
 ### License