aljaz/dev-gpt
1
0
Fork 0
mirror of https://github.com/aljazceru/dev-gpt.git synced 2025-12-18 14:14:21 +01:00
Code Issues Packages Projects Releases Wiki Activity

fix: multiple things and 3d

Browse Source
This commit is contained in:
Florian Hönicke
2023-04-18 02:25:28 +02:00
parent badf295f71
commit 7dcd690245
12 changed files with 176 additions and 129 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
README.md
View File

@@ -103,51 +103,57 @@ jc delete <microservice id>
### Animal Detector
```bash
gptdeploy generate --description "Given an image, return the image with bounding boxes of all animals (https://pjreddie.com/media/files/yolov3.weights, https://raw.githubusercontent.com/pjreddie/darknet/master/cfg/yolov3.cfg)" --test "https://images.unsplash.com/photo-1444212477490-ca407925329e contains animals" --model gpt-4
gptdeploy generate --description "Given an image, return the image with bounding boxes of all animals (https://pjreddie.com/media/files/yolov3.weights, https://raw.githubusercontent.com/pjreddie/darknet/master/cfg/yolov3.cfg)" --test "https://images.unsplash.com/photo-1444212477490-ca407925329e contains animals" --model gpt-4 --path microservice
```
<img src="res/animal_detector_example.png" alt="Animal Detector" width="600" />
### Meme Generator
```bash
gptdeploy generate --description "Generate a meme from an image and a caption" --test "Surprised Pikachu: https://media.wired.com/photos/5f87340d114b38fa1f8339f9/master/w_1600%2Cc_limit/Ideas_Surprised_Pikachu_HD.jpg, TOP:When you discovered GPTDeploy" --model gpt-4
gptdeploy generate --description "Generate a meme from an image and a caption" --test "Surprised Pikachu: https://media.wired.com/photos/5f87340d114b38fa1f8339f9/master/w_1600%2Cc_limit/Ideas_Surprised_Pikachu_HD.jpg, TOP:When you discovered GPTDeploy" --model gpt-4 --path microservice
```
<img src="res/meme_example.png" alt="Meme Generator" width="600" />
### Rhyme Generator
```bash
gptdeploy generate --description "Given a word, return a list of rhyming words using the datamuse api" --test "hello" --model gpt-4
gptdeploy generate --description "Given a word, return a list of rhyming words using the datamuse api" --test "hello" --model gpt-4 --path microservice
```
<img src="res/rhyme_generator_example.png" alt="Rhyme Generator" width="600" />
### Word Cloud Generator
```bash
gptdeploy generate --description "Generate a word cloud from a given text" --test "Lorem ipsum dolor sit amet, consectetur adipiscing elit." --model gpt-4
gptdeploy generate --description "Generate a word cloud from a given text" --test "Lorem ipsum dolor sit amet, consectetur adipiscing elit." --model gpt-4 --path microservice
```
<img src="res/word_cloud_example.png" alt="Word Cloud Generator" width="600" />
### 3d model info
```bash
gptdeploy generate --description "Given a 3d object, return vertex count and face count" --test "https://raw.githubusercontent.com/polygonjs/polygonjs-assets/master/models/wolf.obj" --model gpt-4
gptdeploy generate --description "Given a 3d object, return vertex count and face count" --test "https://raw.githubusercontent.com/polygonjs/polygonjs-assets/master/models/wolf.obj" --model gpt-4 --path microservice
```
<img src="res/obj_info_example.png" alt="3D Model Info" width="600" />
### 2d rendering of 3d model
```bash
gptdeploy generate --description "create a 2d rendering of a whole 3d object and x,y,z object rotation using trimesh and pyrender.OffscreenRenderer with os.environ['PYOPENGL_PLATFORM'] = 'egl' and freeglut3-dev library" --test "input: https://graphics.stanford.edu/courses/cs148-10-summer/as3/code/as3/teapot.obj output: assert the image is not completely white or black" --model gpt-4 --path microservice
```
<img src="res/obj_render_example.gif" alt="2D Rendering of 3D Model" width="600" />
### Table extraction
```bash
gptdeploy generate --description "Given a URL, extract all tables as csv" --test "http://www.ins.tn/statistiques/90" --model gpt-4
gptdeploy generate --description "Given a URL, extract all tables as csv" --test "http://www.ins.tn/statistiques/90" --model gpt-4 --path microservice
```
<img src="res/table_extraction_example.png" alt="Table Extraction" width="600" />
### Audio to mel spectrogram
```bash
gptdeploy generate --description "Create mel spectrograms from audio file" --test "https://cdn.pixabay.com/download/audio/2023/02/28/audio_550d815fa5.mp3" --model gpt-4
gptdeploy generate --description "Create mel spectrograms from audio file" --test "https://cdn.pixabay.com/download/audio/2023/02/28/audio_550d815fa5.mp3" --model gpt-4 --path microservice
```
<img src="res/audio_to_mel_example.png" alt="Audio to Mel Spectrogram" width="600" />
### Text to speech
```bash
gptdeploy generate --description "Convert text to speech" --test "Hello, welcome to GPT Deploy!" --model gpt-4
gptdeploy generate --description "Convert text to speech" --test "Hello, welcome to GPT Deploy!" --model gpt-4 --path microservice
```
<a href=res/text_to_speech_example.wav><img src="res/text_to_speech_example.png" alt="Text to Speech" width="600" /></a>
@@ -158,20 +164,20 @@ gptdeploy generate --description "Convert text to speech" --test "Hello, welcome
### Heatmap Generator
```bash
gptdeploy generate --description "Create a heatmap from an image and a list of relative coordinates" --test "https://images.unsplash.com/photo-1574786198875-49f5d09fe2d2, [[0.1, 0.2], [0.3, 0.4], [0.5, 0.6], [0.2, 0.1], [0.7, 0.2], [0.4, 0.2]]" --model gpt-4
gptdeploy generate --description "Create a heatmap from an image and a list of relative coordinates" --test "https://images.unsplash.com/photo-1574786198875-49f5d09fe2d2, [[0.1, 0.2], [0.3, 0.4], [0.5, 0.6], [0.2, 0.1], [0.7, 0.2], [0.4, 0.2]]" --model gpt-4 --path microservice
```
<img src="res/heatmap_example.png" alt="Heatmap Generator" width="600" />
### QR Code Generator
```bash
gptdeploy generate --description "Generate QR code from URL" --test "https://www.example.com" --model gpt-4
gptdeploy generate --description "Generate QR code from URL" --test "https://www.example.com" --model gpt-4 --path microservice
```
<img src="res/qr_example.png" alt="QR Code Generator" width="600" />
### Mandelbrot Set Visualizer
```bash
gptdeploy generate --description "Visualize the Mandelbrot set with custom parameters" --test "center=-0+1i, zoom=1.0, size=800x800, iterations=1000" --model gpt-4
gptdeploy generate --description "Visualize the Mandelbrot set with custom parameters" --test "center=-0+1i, zoom=1.0, size=800x800, iterations=1000" --model gpt-4 --path microservice
```
<img src="res/mandelbrot_example.png" alt="Mandelbrot Set Visualizer" width="600" />

BIN
res/obj_render_example.gif Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 898 KiB

14
src/apis/gpt.py
View File

@@ -14,7 +14,9 @@ from src.utils.string_tools import print_colored
class GPTSession:
def __init__(self, model: str = 'gpt-4'):
def __init__(self, task_description, test_description, model: str = 'gpt-4', ):
self.task_description = task_description
self.test_description = test_description
self.configure_openai_api_key()
if model == 'gpt-4' and self.is_gpt4_available():
self.supported_model = 'gpt-4'
@@ -68,18 +70,18 @@ If you have updated it already, please restart your terminal.
print('\n')
def get_conversation(self, system_definition_examples: List[str] = ['executor', 'docarray', 'client']):
return _GPTConversation(self.supported_model, self.cost_callback, system_definition_examples)
return _GPTConversation(self.supported_model, self.cost_callback, self.task_description, self.test_description, system_definition_examples)
def calculate_money_spent(self, num_chars, price):
return round(num_chars / CHARS_PER_TOKEN * price / 1000, 3)
class _GPTConversation:
def __init__(self, model: str, cost_callback, system_definition_examples: List[str] = ['executor', 'docarray', 'client']):
def __init__(self, model: str, cost_callback, task_description, test_description, system_definition_examples: List[str] = ['executor', 'docarray', 'client']):
self.model = model
self.cost_callback = cost_callback
self.prompt_list: List[Optional[Tuple]] = [None]
self.set_system_definition(system_definition_examples)
self.set_system_definition(task_description, test_description, system_definition_examples)
if os.environ['VERBOSE'].lower() == 'true':
print_colored('system', self.prompt_list[0][1], 'magenta')
@@ -91,8 +93,8 @@ class _GPTConversation:
self.prompt_list.append(('assistant', response))
return response
def set_system_definition(self, system_definition_examples: List[str] = []):
system_message = system_base_definition
def set_system_definition(self, task_description, test_description, system_definition_examples: List[str] = []):
system_message = system_base_definition(task_description, test_description)
if 'executor' in system_definition_examples:
system_message += f'\n{executor_example}'
if 'docarray' in system_definition_examples:

27
src/apis/jina_cloud.py
View File

@@ -248,10 +248,11 @@ def update_client_line_in_file(file_path, host):
file.write(replaced_content)
def remove_after_stderr(relevant_lines):
def shorten_logs(relevant_lines):
for index, line in enumerate(relevant_lines):
if '--- Captured stderr call ----' in line:
return relevant_lines[:index]
relevant_lines = relevant_lines[:index]
relevant_lines = [line for line in relevant_lines if ' Requirement already satisfied: ' not in line]
return relevant_lines
@@ -270,9 +271,9 @@ def process_error_message(error_message):
if last_matching_line_index is not None:
relevant_lines = lines[last_matching_line_index:]
relevant_lines = remove_after_stderr(relevant_lines)
relevant_lines = shorten_logs(relevant_lines)
response = '\n'.join(relevant_lines[-25:]).strip()
response = '\n'.join(relevant_lines[-100:]).strip()
# the following code tests the case that the docker file is corrupted and can not be parsed
# the method above will not return a relevant error message in this case
@@ -282,21 +283,3 @@ def process_error_message(error_message):
if not response and last_line.startswith('error: '):
return last_line
return response
def build_docker(path):
# The command to build the Docker image
cmd = f"docker build -t micromagic {path}"
# Run the command and capture the output
process = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, shell=True)
stdout, stderr = process.communicate()
# Check if there was an error
if process.returncode != 0:
error_message = stderr.decode("utf-8")
relevant_error_message = process_error_message(error_message)
return relevant_error_message
else:
print("Docker build completed successfully.")
return ''

4
src/cli.py
View File

@@ -63,8 +63,8 @@ def generate(
return
from src.options.generate.generator import Generator
generator = Generator(model=model)
generator.generate(description, test, path)
generator = Generator(description, test, model=model)
generator.generate(path)
@main.command()
@path_param

3
src/constants.py
View File

@@ -36,5 +36,6 @@ MAX_DEBUGGING_ITERATIONS = 10
DEMO_TOKEN = '45372338e04f5a41af949024db929d46'
PROBLEMATIC_PACKAGES = [
'Pyrender', 'Trimesh', 'ModernGL', 'PyOpenGL', 'Pyglet', 'pythreejs', 'panda3d' # because they need a screen
# 'Pyrender', 'Trimesh',
'ModernGL', 'PyOpenGL', 'Pyglet', 'pythreejs', 'panda3d' # because they need a screen
]

12
src/options/__init__.py
View File

@@ -1,13 +1,19 @@
import os
def get_latest_folder(path):
return max([os.path.join(path, f) for f in os.listdir(path) if os.path.isdir(os.path.join(path, f))])
def get_latest_folder(path, max_fn=max):
return max_fn([os.path.join(path, f) for f in os.listdir(path) if os.path.isdir(os.path.join(path, f))])
def version_max_fn(path_list):
version_list = [int(path.split('/')[-1].replace('v', '')) for path in path_list]
max_version = max(version_list)
max_index = version_list.index(max_version)
return path_list[max_index]
def get_latest_version_path(microservice_path):
executor_name_path = get_latest_folder(microservice_path)
latest_approach_path = get_latest_folder(executor_name_path)
latest_version_path = get_latest_folder(latest_approach_path)
latest_version_path = get_latest_folder(latest_approach_path, max_fn=version_max_fn)
return latest_version_path
def get_executor_name(microservice_path):

157
src/options/generate/generator.py
View File

@@ -14,22 +14,23 @@ from src.utils.string_tools import print_colored
class Generator:
def __init__(self, model='gpt-4'):
self.gpt_session = gpt.GPTSession(model=model)
def __init__(self, task_description, test_description, model='gpt-4'):
self.gpt_session = gpt.GPTSession(task_description, test_description, model=model)
self.task_description = task_description
self.test_description = test_description
def extract_content_from_result(self, plain_text, file_name, match_single_block=False):
pattern = fr"^\*\*{file_name}\*\*\n```(?:\w+\n)?([\s\S]*?)```"
match = re.search(pattern, plain_text, re.MULTILINE)
if match:
return match.group(1).strip()
else:
elif match_single_block:
# Check for a single code block
single_code_block_pattern = r"^```(?:\w+\n)?([\s\S]*?)```"
single_code_block_match = re.findall(single_code_block_pattern, plain_text, re.MULTILINE)
if match_single_block and len(single_code_block_match) == 1:
if len(single_code_block_match) == 1:
return single_code_block_match[0].strip()
else:
return ''
return ''
def write_config_yml(self, microservice_name, dest_folder):
config_content = f'''
@@ -61,24 +62,19 @@ class Generator:
test,
path,
microservice_name,
package,
packages,
num_approach,
is_chain_of_thought=False,
):
MICROSERVICE_FOLDER_v1 = get_microservice_path(path, microservice_name, package, num_approach, 1)
MICROSERVICE_FOLDER_v1 = get_microservice_path(path, microservice_name, packages, num_approach, 1)
os.makedirs(MICROSERVICE_FOLDER_v1)
print_colored('', '############# Microservice #############', 'blue')
user_query = (
general_guidelines()
+ executor_file_task(microservice_name, description, test, package)
+ executor_file_task(microservice_name, description, test, packages)
)
conversation = self.gpt_session.get_conversation()
microservice_content_raw = conversation.query(user_query)
if is_chain_of_thought:
microservice_content_raw = conversation.query(
f"General rules: " + not_allowed_executor() + chain_of_thought_optimization('python',
'microservice.py'))
microservice_content = self.extract_content_from_result(microservice_content_raw, 'microservice.py',
match_single_block=True)
if microservice_content == '':
@@ -96,12 +92,6 @@ class Generator:
)
conversation = self.gpt_session.get_conversation()
test_microservice_content_raw = conversation.query(user_query)
if is_chain_of_thought:
test_microservice_content_raw = conversation.query(
f"General rules: " + not_allowed_executor() +
chain_of_thought_optimization('python', 'test_microservice.py')
+ "Don't add any additional tests. "
)
test_microservice_content = self.extract_content_from_result(
test_microservice_content_raw, 'microservice.py', match_single_block=True
)
@@ -117,9 +107,6 @@ class Generator:
)
conversation = self.gpt_session.get_conversation()
requirements_content_raw = conversation.query(user_query)
if is_chain_of_thought:
requirements_content_raw = conversation.query(
chain_of_thought_optimization('', requirements_path) + "Keep the same version of jina ")
requirements_content = self.extract_content_from_result(requirements_content_raw, 'requirements.txt',
match_single_block=True)
@@ -135,9 +122,6 @@ class Generator:
)
conversation = self.gpt_session.get_conversation()
dockerfile_content_raw = conversation.query(user_query)
if is_chain_of_thought:
dockerfile_content_raw = conversation.query(
f"General rules: " + not_allowed_executor() + chain_of_thought_optimization('dockerfile', 'Dockerfile'))
dockerfile_content = self.extract_content_from_result(dockerfile_content_raw, 'Dockerfile',
match_single_block=True)
persist_file(dockerfile_content, os.path.join(MICROSERVICE_FOLDER_v1, 'Dockerfile'))
@@ -167,7 +151,7 @@ print(response[0].text) # can also be blob in case of image/audio..., this shoul
```
Note that the response will always be in response[0].text
You must provide the complete file with the exact same syntax to wrap the code.
The playground (app.py) must read the host from sys.argv because it will be started with a custom host: streamlit run app.py -- --host grpc://...
The playground (app.py) must read the host from sys.argv[-1] because it will be started with a custom host: streamlit run app.py -- --host grpc://...
The playground (app.py) must not let the user configure the host on the ui.
'''
)
@@ -178,7 +162,6 @@ The playground (app.py) must not let the user configure the host on the ui.
persist_file(playground_content, os.path.join(microservice_path, 'app.py'))
def debug_microservice(self, path, microservice_name, num_approach, packages, description, test):
error_before = ''
for i in range(1, MAX_DEBUGGING_ITERATIONS):
print('Debugging iteration', i)
print('Trying to build the microservice. Might take a while...')
@@ -188,29 +171,31 @@ The playground (app.py) must not let the user configure the host on the ui.
error = process_error_message(log_hubble)
if error:
print('An error occurred during the build process. Feeding the error back to the assistent...')
self.do_debug_iteration(description, error, error_before, next_microservice_path,
self.do_debug_iteration(description, error, next_microservice_path,
previous_microservice_path, test)
error_before = error
if i == MAX_DEBUGGING_ITERATIONS - 1:
raise self.MaxDebugTimeReachedException('Could not debug the microservice.')
else:
print('Successfully build microservice.')
break
if i == MAX_DEBUGGING_ITERATIONS - 1:
raise self.MaxDebugTimeReachedException('Could not debug the microservice.')
return get_microservice_path(path, microservice_name, packages, num_approach, i)
def do_debug_iteration(self, description, error, error_before, next_microservice_path, previous_microservice_path,
def do_debug_iteration(self, description, error, next_microservice_path, previous_microservice_path,
test):
os.makedirs(next_microservice_path)
file_name_to_content = get_all_microservice_files_with_content(previous_microservice_path)
summarized_error = self.summarize_error(error)
is_dependency_issue = self.is_dependency_issue(error, file_name_to_content['Dockerfile'])
if is_dependency_issue:
all_files_string = self.files_to_string({
key: val for key, val in file_name_to_content.items() if
key in ['requirements.txt', 'Dockerfile']
})
user_query = self.get_user_query_dependency_issue(all_files_string, error)
user_query = self.get_user_query_dependency_issue(all_files_string, summarized_error)
else:
user_query = self.get_user_query_code_issue(description, error, file_name_to_content,
user_query = self.get_user_query_code_issue(description, summarized_error, file_name_to_content,
test)
conversation = self.gpt_session.get_conversation()
returned_files_raw = conversation.query(user_query)
@@ -218,33 +203,50 @@ The playground (app.py) must not let the user configure the host on the ui.
updated_file = self.extract_content_from_result(returned_files_raw, file_name)
if updated_file and (not is_dependency_issue or file_name in ['requirements.txt', 'Dockerfile']):
file_name_to_content[file_name] = updated_file
print(f'Updated {file_name}')
for file_name, content in file_name_to_content.items():
persist_file(content, os.path.join(next_microservice_path, file_name))
def get_user_query_dependency_issue(self, all_files_string, error):
def get_user_query_dependency_issue(self, all_files_string, summarized_error):
user_query = (
f'''
Your task is to provide guidance on how to solve an error that occurred during the Docker build process.
The error message is:
**microservice.log**
```
{error}
```
Here is the summary of the error that occurred:
{summarized_error}
To solve this error, you should:
1. Identify the type of error by examining the stack trace.
2. Suggest how to solve it.
3. Your suggestion must include the files that need to be changed, but not files that don't need to be changed.
1. Suggest 3 to 5 possible solutions on how to solve it. You have no access to the documentation of the package.
2. Decide for the best solution and explain it in detail.
3. Write down the files that need to be changed, but not files that don't need to be changed.
For files that need to be changed, you must provide the complete file with the exact same syntax to wrap the code.
Obey the following rules: {not_allowed_docker()}
You are given the following files:
{all_files_string}"
Output all the files that need change.
Don't output files that don't need change. If you output a file, then write the
complete file. Use the exact following syntax to wrap the code:
**...**
```
...code...
```
Example:
**requirements.txt**
```
jina==2.0.0
```
'''
)
return user_query
def get_user_query_code_issue(self, description, error, file_name_to_content, test):
def get_user_query_code_issue(self, description, summarized_error, file_name_to_content, test):
all_files_string = self.files_to_string(file_name_to_content)
return f'''
General rules: {not_allowed_executor()}
@@ -256,18 +258,34 @@ Here are all the files I use:
{all_files_string}
This is the error I encounter currently during the docker build process:
{error}
Here is the summary of the error that occurred:
{summarized_error}
Look at the stack trace of the current error. First, think about what kind of error is this?
Then think about possible reasons which might have caused it. Then suggest how to
solve it. Output all the files that need change.
To solve this error, you should:
1. Suggest 3 to 5 possible solutions on how to solve it. You have no access to the documentation of the package.
2. Decide for the best solution and explain it in detail.
3. Write down the files that need to be changed, but not files that don't need to be changed.
Obey the following rules:
{not_allowed_executor()}
{not_allowed_docker()}
Output all the files that need change.
Don't output files that don't need change. If you output a file, then write the
complete file. Use the exact same syntax to wrap the code:
complete file. Use the exact following syntax to wrap the code:
**...**
```...
...code...
```
Example:
**microservice.py**
```python
print('hello world')
```
'''
class MaxDebugTimeReachedException(BaseException):
@@ -332,32 +350,32 @@ If the package is mentioned in the description, then it is automatically the bes
The output must be a list of lists wrapped into ``` and starting with **packages.csv** like this:
**packages.csv**
```
package1
package2
package3
package4
package5
package1a, package1b ...
package2a, package2b, package2c
package3a ...
package4a ...
package5a ...
...
```
'''
conversation = self.gpt_session.get_conversation()
packages_raw = conversation.query(user_query)
packages_csv_string = self.extract_content_from_result(packages_raw, 'packages.csv')
packages = [package.split(',') for package in packages_csv_string.split('\n')]
packages = packages[:NUM_IMPLEMENTATION_STRATEGIES]
return packages
packages_list = [[pkg.strip() for pkg in packages_string.split(',')] for packages_string in packages_csv_string.split('\n')]
packages_list = packages_list[:NUM_IMPLEMENTATION_STRATEGIES]
return packages_list
def generate(self, description, test, microservice_path):
generated_name = self.generate_microservice_name(description)
def generate(self, microservice_path):
generated_name = self.generate_microservice_name(self.task_description)
microservice_name = f'{generated_name}{random.randint(0, 10_000_000)}'
packages_list = self.get_possible_packages(description)
packages_list = self.get_possible_packages(self.task_description)
packages_list = [packages for packages in packages_list if len(set(packages).intersection(set(PROBLEMATIC_PACKAGES))) == 0]
for num_approach, packages in enumerate(packages_list):
try:
self.generate_microservice(description, test, microservice_path, microservice_name, packages,
self.generate_microservice(self.task_description, self.test_description, microservice_path, microservice_name, packages,
num_approach)
final_version_path = self.debug_microservice(microservice_path, microservice_name, num_approach,
packages, description, test)
packages, self.task_description, self.test_description)
self.generate_playground(microservice_name, final_version_path)
except self.MaxDebugTimeReachedException:
print('Could not debug the Microservice with the approach:', packages)
@@ -373,3 +391,14 @@ gptdeploy deploy --path {microservice_path}
'''
)
break
def summarize_error(self, error):
conversation = self.gpt_session.get_conversation([])
user_query = f'''
Here is an error message I encountered during the docker build process:
"{error}"
Your task is to summarize the error message as compact and informative as possible while maintaining all information necessary to debug the core issue.
Warnings are not worth mentioning.
'''
error_summary = conversation.query(user_query)
return error_summary

23
src/options/generate/prompt_system.py
View File

@@ -1,4 +1,5 @@
from src.constants import FLOW_URL_PLACEHOLDER
from src.options.generate.prompt_tasks import not_allowed_executor, not_allowed_docker
executor_example = '''Using the Jina framework, users can define executors.
Here is an example of how an executor can be defined. It always starts with a comment:
@@ -16,7 +17,7 @@ class MyInfoExecutor(Executor):
for d in docs:
content = json.loads(d.text)
...
d.text = json.dumps(modified_content)
d.text = json.dumps(modified_content) # serialized json
return docs
```
@@ -28,7 +29,7 @@ A Document is a python class that represents a single document.
Here is the protobuf definition of a Document:
```
message DocumentProto {{
// used to store json data the executor gets and returns
// used to store serialized json data the executor gets and returns
string text = 1;
}}
```
@@ -71,8 +72,22 @@ print(response[0].text)
```'''
system_base_definition = f'''
def system_base_definition(task_description, test_description):
return f'''
It is the year 2021.
You are a principal engineer working at Jina - an open source company.
You are a principal engineer working at Jina - an open source company.
You accurately satisfy all of the user's requirements.
To be more specific, you help the user to build a microservice with the following requirements:
```
{task_description}
```
and the following test scenario:
```
{test_description}
```
You must obey the following rules:
{not_allowed_executor()}
{not_allowed_docker()}
'''

29
src/options/generate/prompt_tasks.py
View File

@@ -4,17 +4,22 @@ from src.constants import EXECUTOR_FILE_NAME, REQUIREMENTS_FILE_NAME, TEST_EXECU
def general_guidelines():
return (
"The code you write is production ready. "
"Every file starts with comments describing what the code is doing before the first import. "
"Comments can only be written within code blocks. "
"Then all imports are listed. "
"It is important to import all modules that could be needed in the Executor code. "
"Always import: "
"from jina import Executor, DocumentArray, Document, requests "
"Start from top-level and then fully implement all methods. "
"\n"
)
return '''
The code you write is production ready.
Every file starts with comments describing what the code is doing before the first import.
Comments can only be written within code blocks.
Then all imports are listed.
It is important to import all modules that could be needed in the Executor code.
Always import:
from jina import Executor, DocumentArray, Document, requests
import json
from io import BytesIO
import requests as req
Start from top-level and then fully implement all methods.
'''
def _task(task, tag_name, file_name, purpose=None):
@@ -44,7 +49,7 @@ Have in mind that d.uri is never a path to a local file. It is always a url.
{not_allowed_executor()}
Your approach:
1. Identify the core challenge when implementing the executor.
2. Think about solutions for these challenges.
2. Think about possible solutions for these challenges.
3. Decide for one of the solutions.
4. Write the code.
''', EXECUTOR_FILE_TAG, EXECUTOR_FILE_NAME)

4
src/utils/io.py
View File

@@ -9,8 +9,8 @@ import docker
from docker import APIClient
def get_microservice_path(path, microservice_name, package, num_approach, version):
package_path = '_'.join(package)
def get_microservice_path(path, microservice_name, packages, num_approach, version):
package_path = '_'.join(packages)
return os.path.join(path, microservice_name, f'{num_approach}_{package_path}', f'v{version}')
def persist_file(file_content, file_path):

4
test/test_generator.py
View File

@@ -25,5 +25,5 @@ def test_generator(tmpdir):
# Use mock.patch as a context manager to replace the original methods with the mocks
with mock.patch("openai.ChatCompletion.create", side_effect=mock_create), \
mock.patch.object(GPTSession, "configure_openai_api_key", side_effect=mock_get_openai_api_key):
generator = Generator()
generator.generate("my description", "my test", str(tmpdir))
generator = Generator("my description", "my test")
generator.generate(str(tmpdir))