update changelog

Author: eff-kay

CHANGELOG.md

@@ -2,6 +2,13 @@
 All notable changes to `slashml-python-client` aka `slashml` will be documented in this file.
 This project adheres to [Semantic Versioning](https://semver.org/).
 
+## 0.1.6 - 2023-12-21
+
+### Added
+- Pin some dependencies
+- Add a requirements flag for model deployment
+
+
 ## 0.1.5 - 2023-09-17
 
 ### Added

examples/deploy_custom_models/scikit-learn/.upm/store.json

@@ -0,0 +1 @@
+{"version":2,"languages":{"python-python3-poetry":{"specfileHash":"e466d5cdb4f0114138d43eb6b027fe33","lockfileHash":"f8a716e5ffc6e8184f35537de7eaa860"}}}

examples/deploy_custom_models/scikit-learn/deploy_scikit_learn_models.py

@@ -16,19 +16,9 @@
 # deploy model
 response = client.deploy(model_name='sk-learn-model', model=my_model)
 
-# wait for it to be deployed
-time.sleep(2)
+# check status
 status = client.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 = client.status(model_version_id=response.id)
-
-    if status.status == 'FAILED':
-        raise Exception('Model deployment failed')
-
 # submit prediction
 model_input = [[0, 0, 0, 0]]
 prediction = client.predict(model_version_id=response.id, model_input=model_input)

examples/deploy_custom_models/scikit-learn/my_model/model/model.py

@@ -33,3 +33,6 @@ def predict(self, model_input: Any) -> Any:
                 model_input
             ).tolist()
         return model_output
+
+
+Model().predict(input)

examples/pipeline_podcast_summary.py

@@ -1,4 +1,4 @@
-from slashml import SpeechToText, TextSummarization, TextToSpeech
+from slashml import SpeechToText, TextSummarization, TextToSpeech, ModelDeployment as Pipeline
 
 # Replace `API_KEY` with your SlasML API token. This example still runs without
 # the API token but usage will be limited
@@ -16,22 +16,28 @@
 transcribe = SpeechToText(api_key=API_KEY)
 summarize = TextSummarization(api_key=API_KEY)
 speechify = TextToSpeech(api_key=API_KEY)
+pipeline = Pipeline(api=API_KEY)
 
 
-response = transcribe.execute(
-    upload_url=uploaded_url, service_provider=service_provider_speech_to_text
-)
+response = pipeline.deploy(blueprint = [transcribe, summarize, speechify], upload_url=uploaded_url)
+
+pipeline.status(response.pipeline_id)
+
+
+# response = transcribe.execute(
+#     upload_url=uploaded_url, service_provider=service_provider_speech_to_text
+# )
 
-print('starting pipeline, the first response might take 10 secs')
-transcribed_text = response.transcription_data.transcription
-print (f"Transcribed Text = {transcribed_text}")
+# print('starting pipeline, the first response might take 10 secs')
+# transcribed_text = response.transcription_data.transcription
+# print (f"Transcribed Text = {transcribed_text}")
 
-response_summarize = summarize.execute(transcribed_text, service_provider_summarize)
+# response_summarize = summarize.execute(transcribed_text, service_provider_summarize)
 
-summary = response_summarize.summarization_data
+# summary = response_summarize.summarization_data
 
-print (f"Summarized Text = {summary}")
+# print (f"Summarized Text = {summary}")
 
-response = speechify.execute(summary, service_provider_text_to_speech)
+# response = speechify.execute(summary, service_provider_text_to_speech)
 
-print (f"\n\n\n You can access the audio file here: {response.audio_url}")
+# print (f"\n\n\n You can access the audio file here: {response.audio_url}")

requires-install.txt

@@ -1,3 +1,3 @@
 requests
 addict
-truss
+truss==0.4.8

setup.py

@@ -18,7 +18,7 @@ def read_req_file(req_type):
 
 setup(
     name="slashml",
-    version="0.1.5",
+    version="0.1.6",
     url="https://slashml.com/",
     author="eff-kay",
     author_email="faiizan14@gmail.com",

slashml/model_deployment.py

@@ -35,7 +35,7 @@ def deploy(self, *, model_name:str, model: str, requirements:t.Optional[list] =
 
             requirements_file_path = 'requirements.txt'
 
-        truss.create(model, 'my_model')
+        truss.create(model, 'my_model', requirements_file=requirements_file_path)
         self.create_tar_gz(folder_path='my_model', tar_gz_filename='my_model.tar.gz')
 
         url = generateURL(self._base_url, "models")