Merge pull request #48 from slashml/model-deployment

deploy model example

Author: eff-kay

CHANGELOG.md

@@ -2,6 +2,11 @@
 All notable changes to `slashml-python-client` aka `slashml` will be documented in this file.
 This project adheres to [Semantic Versioning](https://semver.org/).
 
+## 0.1.4 - 2023-06-09
+
+### Added
+- Model deployment
+
 ## 0.1.3 - 2023-05-13
 
 ### Added

DEVELOPMENT.md

@@ -0,0 +1,48 @@
+# Contributing to this library
+
+
+In order to contribute you need to install a few things
+
+
+1. First make sure you have python3 installed. You can check by running the following command in your terminal
+
+```
+python3 -m pip install --upgrade pip
+```
+
+2. Install the build packages
+```
+python3 -m pip install --upgrade build
+```
+
+3. Then run the following command at the root of the project to build the package
+```
+SLASHML_DIR='.' python3 -m build
+```
+
+The `SLASHML_DIR` will pick the root dir and build the package inside the dist folder in the root.
+
+
+4. Pushing the package to pypi
+
+```
+python3 -m pip install --upgrade twine
+python3 -m pip install --upgrade twine
+python3 -m twine upload --repository testpypi dist/*
+```
+
+Make sure dist only has the newly build package, otherwise the command will try to deploy all of the packages.
+
+The above command will prompt you for a username and a password. You will need developer access to the pypi repo to be able to push the package. Contact @eff-kay for this.
+
+
+5. Test the package by 
+
+```
+python -m pip install --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple slashml
+```
+
+This command will pull the latest from test.pypi.org and the extra-index-url will pull the dependencies from pypi.org. This is because the slashml package depends on other packages that are not on test.pypi.org.
+
+
+

README.md

@@ -65,6 +65,56 @@ print(f"Summary = {response.summarization_data}")
 ```
 
 
+#### Deploy your own Model
+<!-- write a code snippet in the minimum number of lines  -->
+
+Note: this examples requires the `transformers` and `torch` packages to be installed. You can install them with `pip install transformers torch`
+
+```
+pip install transformers torch
+```
+
+```python
+from slashml import ModelDeployment
+import time
+
+# you might have to install transfomers and torch
+from transformers import pipeline
+
+def train_model():
+    # Bring in model from huggingface
+    return pipeline('fill-mask', model='bert-base-uncased')
+
+my_model = train_model()
+
+# Replace `API_KEY` with your SlasML API token.
+API_KEY = "YOUR_API_KEY"
+
+model = ModelDeployment(api_key=API_KEY)
+
+# deploy model
+response = model.deploy(model_name='my_model_3', model=my_model)
+
+# wait for it to be deployed
+time.sleep(2)
+status = model.status(model_version_id=response.id)
+
+while status.status != 'READY':
+    print(f'status: {status.status}')
+    print('trying again in 5 seconds')
+    time.sleep(5)
+    status = model.status(model_version_id=response.id)
+
+    if status.status == 'FAILED':
+        raise Exception('Model deployment failed')
+
+# submit prediction
+input_text = 'Steve jobs is the [MASK] of Apple.'
+prediction = model.predict(model_version_id=response.id, model_input=input_text)
+print(prediction)
+```
+
+
 ### View the list of service providers available
 ```python
 from slashml import TextToSpeech

examples/deploy_model/hugging_face_transformer/deploy_hugging_face_transformer.py

@@ -0,0 +1,35 @@
+from slashml import ModelDeployment
+import time
+
+# you might have to install transfomers and torch
+from transformers import pipeline
+
+def train_model():
+    # Bring in model from huggingface
+    return pipeline('fill-mask', model='bert-base-uncased')
+
+my_model = train_model()
+
+# Replace `API_KEY` with your SlasML API token.
+model = ModelDeployment(api_key=None)
+
+# deploy model
+response = model.deploy(model_name='my_model_3', model=my_model)
+
+# wait for it to be deployed
+time.sleep(2)
+status = model.status(model_version_id=response.id)
+
+while status.status != 'READY':
+    print(f'status: {status.status}')
+    print('trying again in 5 seconds')
+    time.sleep(5)
+    status = model.status(model_version_id=response.id)
+
+    if status.status == 'FAILED':
+        raise Exception('Model deployment failed')
+
+# submit prediction
+input_text = 'Steve jobs is the [MASK] of Apple.'
+prediction = model.predict(model_version_id=response.id, model_input=input_text)
+print(prediction)

examples/deploy_model/hugging_face_transformer/requirements.txt

@@ -0,0 +1,3 @@
+transformers==4.30.0
+torch==2.0.1
+slashml==0.1.4

examples/speech_to_text_sync.py

@@ -3,7 +3,8 @@
 # Replace `API_KEY` with your SlasML API token. This example still runs without
 # the API token but usage will be limited
 API_KEY = "YOUR_API_KEY"
-service_provider = SpeechToText.ServiceProvider.REV
+
+service_provider = SpeechToText.ServiceProvider.WHISPER
 audio_filepath = "test.mp3"
 
 # Find all the service providers that we support by running the choices() method

examples/summarize_sync.py

@@ -3,7 +3,7 @@
 
 # Replace `API_KEY` with your SlasML API token. This example still runs without
 # the API token but usage will be limited
-API_KEY = "YOUR_API_KEY"
+API_KEY = None
 service_provider = TextSummarization.ServiceProvider.OPENAI
 
 input_text = """A good writer doesn't just think, and then write down what he thought, as a sort of transcript. A good writer will almost always discover new things in the process of writing. And there is, as far as I know, no substitute for this kind of discovery. Talking about your ideas with other people is a good way to develop them. But even after doing this, you'll find you still discover new things when you sit down to write. There is a kind of thinking that can only be done by writing."""

examples/text_to_speech_sync.py

@@ -3,6 +3,7 @@
 # Replace `API_KEY` with your SlasML API token. This example still runs without
 # the API token but usage will be limited
 API_KEY = "YOUR_API_KEY"
+
 service_provider = TextToSpeech.ServiceProvider.AWS
 input_text = "To be or not to be, that is the question!"
 
@@ -15,4 +16,4 @@
 # Submit request
 job = model.execute(text=input_text, service_provider=service_provider)
 
-print (f"\n\n\n You can access the audio file here: {job.audio_url}")
+print(f"\n\n\n You can access the audio file here: {job.audio_url}")

requires-build.txt

@@ -0,0 +1,3 @@
+build==0.10.0
+twine==4.0.2
+

requires-install.txt

@@ -1,2 +1,3 @@
 requests==2.28.1
 addict==2.4.0
+truss==0.4.8