xiaxinhang/pr-agent
1
0
Fork 0
mirror of https://github.com/qodo-ai/pr-agent.git synced 2025-07-04 12:50:38 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #1387 from KennyDizi/main

Browse Source 
Introduce to weak model
This commit is contained in:
Tal
2024-12-11 17:36:18 +02:00
committed by GitHub
parent da786b8020 f9a7b18073
commit cb88489dbe
12 changed files with 26 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
docs/docs/usage-guide/changing_a_model.md
View File

@ -5,7 +5,7 @@ To use a different model than the default (GPT-4), you need to edit in the [conf
```
[config]
model = "..."
model_turbo = "..."
model_weak = "..."
fallback_models = ["..."]
```
@ -28,7 +28,7 @@ and set in your configuration file:
```
[config]
model="" # the OpenAI model you've deployed on Azure (e.g. gpt-3.5-turbo)
model_turbo="" # the OpenAI model you've deployed on Azure (e.g. gpt-3.5-turbo)
model_weak="" # the OpenAI model you've deployed on Azure (e.g. gpt-3.5-turbo)
fallback_models=["..."] # the OpenAI model you've deployed on Azure (e.g. gpt-3.5-turbo)
```
@ -52,7 +52,7 @@ MAX_TOKENS={
[config] # in configuration.toml
model = "ollama/llama2"
model_turbo = "ollama/llama2"
model_weak = "ollama/llama2"
fallback_models=["ollama/llama2"]
[ollama] # in .secrets.toml
@ -76,7 +76,7 @@ MAX_TOKENS={
}
[config] # in configuration.toml
model = "huggingface/meta-llama/Llama-2-7b-chat-hf"
model_turbo = "huggingface/meta-llama/Llama-2-7b-chat-hf"
model_weak = "huggingface/meta-llama/Llama-2-7b-chat-hf"
fallback_models=["huggingface/meta-llama/Llama-2-7b-chat-hf"]
[huggingface] # in .secrets.toml
@ -91,7 +91,7 @@ To use Llama2 model with Replicate, for example, set:
```
[config] # in configuration.toml
model = "replicate/llama-2-70b-chat:2c1608e18606fad2812020dc541930f2d0495ce32eee50074220b87300bc16e1"
model_turbo = "replicate/llama-2-70b-chat:2c1608e18606fad2812020dc541930f2d0495ce32eee50074220b87300bc16e1"
model_weak = "replicate/llama-2-70b-chat:2c1608e18606fad2812020dc541930f2d0495ce32eee50074220b87300bc16e1"
fallback_models=["replicate/llama-2-70b-chat:2c1608e18606fad2812020dc541930f2d0495ce32eee50074220b87300bc16e1"]
[replicate] # in .secrets.toml
key = ...
@ -107,7 +107,7 @@ To use Llama3 model with Groq, for example, set:
```
[config] # in configuration.toml
model = "llama3-70b-8192"
model_turbo = "llama3-70b-8192"
model_weak = "llama3-70b-8192"
fallback_models = ["groq/llama3-70b-8192"]
[groq] # in .secrets.toml
key = ... # your Groq api key
@ -121,7 +121,7 @@ To use Google's Vertex AI platform and its associated models (chat-bison/codecha
```
[config] # in configuration.toml
model = "vertex_ai/codechat-bison"
model_turbo = "vertex_ai/codechat-bison"
model_weak = "vertex_ai/codechat-bison"
fallback_models="vertex_ai/codechat-bison"
[vertexai] # in .secrets.toml
@ -140,7 +140,7 @@ To use [Google AI Studio](https://aistudio.google.com/) models, set the relevant
```toml
[config] # in configuration.toml
model="google_ai_studio/gemini-1.5-flash"
model_turbo="google_ai_studio/gemini-1.5-flash"
model_weak="google_ai_studio/gemini-1.5-flash"
fallback_models=["google_ai_studio/gemini-1.5-flash"]
[google_ai_studio] # in .secrets.toml
@ -156,7 +156,7 @@ To use Anthropic models, set the relevant models in the configuration section of
```
[config]
model="anthropic/claude-3-opus-20240229"
model_turbo="anthropic/claude-3-opus-20240229"
model_weak="anthropic/claude-3-opus-20240229"
fallback_models=["anthropic/claude-3-opus-20240229"]
```
@ -173,7 +173,7 @@ To use Amazon Bedrock and its foundational models, add the below configuration:
```
[config] # in configuration.toml
model="bedrock/anthropic.claude-3-sonnet-20240229-v1:0"
model_turbo="bedrock/anthropic.claude-3-sonnet-20240229-v1:0"
model_weak="bedrock/anthropic.claude-3-sonnet-20240229-v1:0"
fallback_models=["bedrock/anthropic.claude-v2:1"]
```
@ -195,7 +195,7 @@ If the relevant model doesn't appear [here](https://github.com/Codium-ai/pr-agen
```
[config]
model="custom_model_name"
model_turbo="custom_model_name"
model_weak="custom_model_name"
fallback_models=["custom_model_name"]
```
(2) Set the maximal tokens for the model:

6
pr_agent/algo/pr_processing.py
View File

@ -333,7 +333,7 @@ def generate_full_patch(convert_hunks_to_line_numbers, file_dict, max_tokens_mod
return total_tokens, patches, remaining_files_list_new, files_in_patch_list
async def retry_with_fallback_models(f: Callable, model_type: ModelType = ModelType.REGULAR):
async def retry_with_fallback_models(f: Callable, model_type: ModelType = ModelType.WEAK):
all_models = _get_all_models(model_type)
all_deployments = _get_all_deployments(all_models)
# try each (model, deployment_id) pair until one is successful, otherwise raise exception
@ -354,8 +354,8 @@ async def retry_with_fallback_models(f: Callable, model_type: ModelType = ModelT
def _get_all_models(model_type: ModelType = ModelType.REGULAR) -> List[str]:
if model_type == ModelType.TURBO:
model = get_settings().config.model_turbo
if get_settings().config.get('model_weak') and model_type == ModelType.WEAK:
model = get_settings().config.model_weak
else:
model = get_settings().config.model
fallback_models = get_settings().config.fallback_models

3
pr_agent/algo/utils.py
View File

@ -35,8 +35,7 @@ class Range(BaseModel):
class ModelType(str, Enum):
REGULAR = "regular"
TURBO = "turbo"
WEAK = "weak"
class PRReviewHeader(str, Enum):
REGULAR = "## PR Reviewer Guide"

2
pr_agent/git_providers/utils.py
View File

@ -99,5 +99,5 @@ def set_claude_model():
"""
model_claude = "bedrock/anthropic.claude-3-5-sonnet-20240620-v1:0"
get_settings().set('config.model', model_claude)
get_settings().set('config.model_turbo', model_claude)
get_settings().set('config.model_weak', model_claude)
get_settings().set('config.fallback_models', [model_claude])

4
pr_agent/settings/configuration.toml
View File

@ -1,7 +1,7 @@
[config]
# models
model="gpt-4-turbo-2024-04-09"
model_turbo="gpt-4o-2024-11-20"
model_weak="gpt-4o-mini-2024-07-18"
model="gpt-4o-2024-11-20"
fallback_models=["gpt-4o-2024-08-06"]
# CLI
git_provider="github"

4
pr_agent/tools/pr_code_suggestions.py
View File

@ -114,9 +114,9 @@ class PRCodeSuggestions:
# call the model to get the suggestions, and self-reflect on them
if not self.is_extended:
data = await retry_with_fallback_models(self._prepare_prediction)
data = await retry_with_fallback_models(self._prepare_prediction, model_type=ModelType.REGULAR)
else:
data = await retry_with_fallback_models(self._prepare_prediction_extended)
data = await retry_with_fallback_models(self._prepare_prediction_extended, model_type=ModelType.REGULAR)
if not data:
data = {"code_suggestions": []}
self.data = data

2
pr_agent/tools/pr_description.py
View File

@ -99,7 +99,7 @@ class PRDescription:
# ticket extraction if exists
await extract_and_cache_pr_tickets(self.git_provider, self.vars)
await retry_with_fallback_models(self._prepare_prediction, ModelType.TURBO)
await retry_with_fallback_models(self._prepare_prediction, ModelType.WEAK)
if self.prediction:
self._prepare_data()

2
pr_agent/tools/pr_help_message.py
View File

@ -114,7 +114,7 @@ class PRHelpMessage:
self.vars['snippets'] = docs_prompt.strip()
# run the AI model
response = await retry_with_fallback_models(self._prepare_prediction, model_type=ModelType.REGULAR)
response = await retry_with_fallback_models(self._prepare_prediction, model_type=ModelType.WEAK)
response_yaml = load_yaml(response)
response_str = response_yaml.get('response')
relevant_sections = response_yaml.get('relevant_sections')

2
pr_agent/tools/pr_line_questions.py
View File

@ -79,7 +79,7 @@ class PR_LineQuestions:
line_end=line_end,
side=side)
if self.patch_with_lines:
response = await retry_with_fallback_models(self._get_prediction, model_type=ModelType.TURBO)
response = await retry_with_fallback_models(self._get_prediction, model_type=ModelType.WEAK)
get_logger().info('Preparing answer...')
if comment_id:

2
pr_agent/tools/pr_questions.py
View File

@ -63,7 +63,7 @@ class PRQuestions:
if img_path:
get_logger().debug(f"Image path identified", artifact=img_path)
await retry_with_fallback_models(self._prepare_prediction, model_type=ModelType.TURBO)
await retry_with_fallback_models(self._prepare_prediction, model_type=ModelType.WEAK)
pr_comment = self._prepare_pr_answer()
get_logger().debug(f"PR output", artifact=pr_comment)

2
pr_agent/tools/pr_reviewer.py
View File

@ -148,7 +148,7 @@ class PRReviewer:
if get_settings().config.publish_output and not get_settings().config.get('is_auto_command', False):
self.git_provider.publish_comment("Preparing review...", is_temporary=True)
await retry_with_fallback_models(self._prepare_prediction)
await retry_with_fallback_models(self._prepare_prediction, model_type=ModelType.REGULAR)
if not self.prediction:
self.git_provider.remove_initial_comment()
return None

2
pr_agent/tools/pr_update_changelog.py
View File

@ -73,7 +73,7 @@ class PRUpdateChangelog:
if get_settings().config.publish_output:
self.git_provider.publish_comment("Preparing changelog updates...", is_temporary=True)
await retry_with_fallback_models(self._prepare_prediction, model_type=ModelType.TURBO)
await retry_with_fallback_models(self._prepare_prediction, model_type=ModelType.WEAK)
new_file_content, answer = self._prepare_changelog_update()