microsoft / pi-azure-recipes Goto Github PK

The language in the tutorial often condenses multiple steps into one paragraph. This is hard to read and should be better organized and broken up into ordered lists.

Deploying the function locally needs updates to ensure it's not broken.
Function executed as expected, but did not see the output as indicated in the instructions. Pulling the URL from instructions did work and generated a 'device is on' message output on the Pi.

Step 3: Suggest rephrasing b/c as-is it's confusing. IoT Hub is at the bottom of the page. Options menu, '...' is on that bottom tab.
Step 8: Unclear what folder folks should look in for the local.settings.json file
Step 11: We don't have a device ID at this stage, that is done in the next section. Suggest moving this to after Device ID is added. (Or making sure that what folks put here is used for the next section)

Errors and clarification for Pi Recipe #2 Prep environment section:
step 4: recipe has been renamed, need to update folder name
step 5: keyboard shortcut doesn't work on windows PC
step 9: before this step, folks are asked about creating a virtual environment and need to indicate 'Yes'

In RAW DATA and Overview - no telemetry data.

In RaspberryPi it only displays:

pi@raspberrypi:~ $ cd client
pi@raspberrypi:/client $ source ./.venv/bin/activate
(04_iot_central) pi@raspberrypi:/client $ python raspberry_pi_client.py
Press Q to quit
Sending reported property: {'remaining_disk_space': '19G'}

Errors are displayed when the program is interrupted by Q.

Press Q to quit
q
Quitting...
Traceback (most recent call last):
File "/home/pi/client/raspberry_pi_client.py", line 296, in
asyncio.run(main())
File "/usr/lib/python3.9/asyncio/runners.py", line 44, in run
return loop.run_until_complete(main)
File "/usr/lib/python3.9/asyncio/base_events.py", line 642, in run_until_complete
return future.result()
File "/home/pi/client/raspberry_pi_clie

nt.py", line 288, in main
await asyncio.gather(telemetry_loop, reported_loop)
File "/home/pi/client/raspberry_pi_client.py", line 161, in send_telemetry
payload = f'{{"cpu_temp": {get_current_cpu_temp()}, "cpu_freq": {get_cpu_frequency_statistics()["current"]}, "cpu_usage": {get_cpu_usage()}, "ram_usage": {get_memory_statistics()["used"]}}}'
File "/home/pi/client/raspberry_pi_client.py", line 91, in get_cpu_usage
return float(cpu_usage)
ValueError: could not convert string to float: ''
Task was destroyed but it is pending!

Please, can you tell me where this error came from ?
"line 91, in get_cpu_usage
return float(cpu_usage)**
ValueError: could not convert string to float: ''"

About, Commands - works ok.

Where possible, we should try to keep the tutorial student in VS Code and use VS Code commands as they are more reliable and easier to explain without screenshots.

• 01_iot
• 02_c2d_messages
• 03_cv
• 04_iot_central

Suggest adding code comments to all sample code in pi recipes b/c it will help folks modify and adapt for custom scenarios.

I got to the step where I ran the "python3 raspberry_pi_client.py" command but I am seeing a KeyError, shown in attached png.

Was unable to successfully deploy to the Pi.

Suggest using the CLI for CV creation: https://docs.microsoft.com/en-us/azure/cognitive-services/cognitive-services-apis-create-account-cli?tabs=windows#prerequisites

The Azure portal did not work as expected for generating a CV resource due to an unexpected permissions issue (was not related to SLC-alt account but rather requires customer to agree to an AI agreement).

For folks using SSH, a camera preview isn't possible. Suggest adding in some commentary around how to handle this (maybe folks could use a desktop viewer over SSH?).