modaic/code-generator-trl-grpo
3
1
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

(no commit message)

Browse Source
This commit is contained in:
Farouk Adeleke
2025-12-24 00:34:33 -08:00
parent 3982ab9745
commit fb1efa3b4a
3 changed files with 4 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
main.py
View File

@@ -5,7 +5,7 @@ class CodeGeneratorConfig(PrecompiledConfig):
model : str = "openai//models/checkpoint-5"
api_base : str = "https://modaic-ai--grpo-demo-serve.modal.run/v1"
max_tokens : int = 10000
temperature : float = 2
temperature : float = 0.7
class CodeGeneration(dspy.Signature):
@@ -32,6 +32,6 @@ class CodeGenerator(PrecompiledProgram):
code_generator = CodeGenerator(CodeGeneratorConfig())
print(code_generator(query="Write a python function that returns the sum of two numbers.").code)
code_generator.push_to_hub("modaic/code-generator-trl-grpo", with_code=True, tag="v2")
code_generator.push_to_hub("modaic/code-generator-trl-grpo", with_code=True, tag="v2.0.1")