AI control plane to schedule, run, debug, observe, and automate AI/ML workloads, agents, and sandboxes on your infra. Community: polyaxon.com/slack/
Joined December 2016
- Tweets 341
- Following 595
- Followers 1,075
- Likes 145
66 Photos and videos
May 25
Polyaxon's new sandbox client:
from polyaxon.client import SandboxClient
sandbox = SandboxClient(project="demo")
sandbox.ping()
result = sandbox.process.exec(["sh", "-lc", "echo ok"])
print(result.stdout)
print(result.stderr)
print(result.exit_code)
1
2
52
May 25
Programmatically :
pty = sandbox.pty.create(command=["sh"], cols=120, rows=40)
ws = sandbox.pty.attach(pty.pty_id, replay_bytes=0)
from polyaxon._pty.sandbox import SandboxPseudoTerminal
exit_code = SandboxPseudoTerminal(ws).start()
14
May 25
Background exec:
bg = sandbox.process.exec_bg(["sh", "-lc", "while true; do date; sleep 1; done"], tag="clock")
print(bg.id)
for chunk in bg.iter_stdout(timeout=10):
print(chunk, end="")
bg.kill()
1
18
May 25
Fs ops:
sandbox.fs.write_text("/tmp/hello.txt", "hello\n")
sandbox.fs.read_text("/tmp/hello.txt")
sandbox.fs.ls("/tmp")
sandbox.fs.upload_file("./local.txt", "/tmp/f.txt")
sandbox.fs.download_file("/tmp/f.txt", "./copy.txt")
10
May 24
Two new CLI commands are coming in beta in the next Polyaxon version:
`polyaxon sandbox` to launch isolated environments
`polyaxon ssh` to attach to workloads with governed remote access
Built for faster debugging loops across AI/ML workloads and agents.
2
46
May 19
We’ve been quiet here. That changes today.
Polyaxon v3 is in active development: an AI control plane to schedule, run, debug, observe, and automate AI/ML workloads, agents, and sandboxes on your infrastructure.
We’ll share the work as it lands.
4
87
7 Dec 2022
As AI models become more prevalent in decision-making processes, it's important to consider the potential vulnerabilities that come with them. Without proper trust, risk, and security management, AI models are at risk of adversarial attacks. #MLOps #XAI #ModelMonitoring
1
7 Dec 2022
In short, AI model management without proper trust, risk, and security management is vulnerable to adversarial attacks. By prioritizing these areas, organizations can better protect their AI models and ensure their accuracy and reliability. #MLOps
1
7 Dec 2022
At Polyaxon we are building a set of open-source tools to help you manage and accelerate the complete life cycle of your deep learning and machine learning. github.com/polyaxon
5 Dec 2022
Implementing MLOps in your organization using automation and integration of key processes can greatly reduce the time and the effort required to build and deploy machine learning models. #MLOps
1
5 Dec 2022
Monitoring and evaluation of machine learning models perfomance is important to know if they are addressing pain points as expected and to identify and fix any potential issues. #modelmonitoring
1
5 Dec 2022
As the volume and complexity of our machine learning workloads continue to increase, it's crucial that the platform can handle this growth. The #MLOps platform needs to be scalable and flexible.