Views
No views yet
| Task Type | Average Improvement | Best Case |
|---|---|---|
| Loop Optimization | 45-65% | 85% |
| Algorithm Complexity | 60-80% | 200x |
| Memory Usage | 30-50% | 75% |
| String Operations | 25-40% | 60% |
1from transformers import AutoTokenizer, AutoModelForCausalLM
2
3tokenizer = AutoTokenizer.from_pretrained("Sairamg18814/jigyasa-agi")
4model = AutoModelForCausalLM.from_pretrained("Sairamg18814/jigyasa-agi")
5
6# Example usage
7code = """
8def find_duplicates(items):
9 duplicates = []
10 for i in range(len(items)):
11 for j in range(i + 1, len(items)):
12 if items[i] == items[j] and items[i] not in duplicates:
13 duplicates.append(items[i])
14 return duplicates
15"""
16
17prompt = f"Optimize this Python function:\n{code}"
18inputs = tokenizer(prompt, return_tensors="pt")
19outputs = model.generate(**inputs, max_length=1000)
20optimized_code = tokenizer.decode(outputs[0], skip_special_tokens=True)1# Install the model
2ollama pull Sairamg18814/jigyasa
3
4# Use for code optimization
5ollama run jigyasa "Optimize this function: def sum_list(items): total = 0; for item in items: total += item; return total"1def find_max(numbers):
2 max_num = numbers[0]
3 for i in range(len(numbers)):
4 if numbers[i] > max_num:
5 max_num = numbers[i]
6 return max_num1def find_max(numbers):
2 return max(numbers) if numbers else None1@software{jigyasa2024,
2 author = {JIGYASA Contributors},
3 title = {JIGYASA: Autonomous General Intelligence for Code Optimization},
4 year = {2024},
5 publisher = {Hugging Face},
6 url = {https://huggingface.co/Sairamg18814/jigyasa-agi}
7}