# Error while using tiny-yolo on prodigy-1.8.3
**URL:** https://support.prodi.gy/t/error-while-using-tiny-yolo-on-prodigy-1-8-3/1692
**Category:** Uncategorized
**Tags:** image
**Created:** [June 21, 2019, 1:32pm UTC](https://support.prodi.gy/t/error-while-using-tiny-yolo-on-prodigy-1-8-3/1692 "2019-06-21T13:32:09Z")
**Posts on this page:** 2
**Page:** 1
### Author:  [@prodigy2019](https://support.prodi.gy/u/prodigy2019)
#### Post date: [June 21, 2019, 1:32pm UTC](https://support.prodi.gy/t/error-while-using-tiny-yolo-on-prodigy-1-8-3/1692/1 "2019-06-21T13:32:09Z")
Hi,
I am new to prodigy and trying to use tiny-yolo for a dataset. I am following cookbook of computer vision - prodigy image.test object\_dataset tiny-yolo /path/to/images\_dir
But this is throwing error -
Exception when serving /get\_session\_questions
Traceback (most recent call last):
File “cython\_src/prodigy/components/feeds.pyx”, line 140, in prodigy.components.feeds.SessionFeed.get\_session\_stream
File “xxx/lib/python3.7/site-packages/toolz/itertoolz.py”, line 368, in first
return next(iter(seq))
StopIteration
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File “xxx/lib/python3.7/site-packages/waitress/channel.py”, line 336, in service
task.service()
File “xxx/lib/python3.7/site-packages/waitress/task.py”, line 175, in service
self.execute()
File “xxx/lib/python3.7/site-packages/waitress/task.py”, line 452, in execute
app\_iter = self.channel.server.application(env, start\_response)
File “xxx/lib/python3.7/site-packages/hug/api.py”, line 451, in api\_auto\_instantiate
return module. **hug\_wsgi** (\*args, \*\*kwargs)
File “xxx/lib/python3.7/site-packages/falcon/api.py”, line 244, in **call**
responder(req, resp, \*\*params)
File “xxx/lib/python3.7/site-packages/hug/interface.py”, line 789, in **call**
raise exception
File “xxx/lib/python3.7/site-packages/hug/interface.py”, line 762, in **call**
self.render\_content(self.call\_function(input\_parameters), context, request, response, \*\*kwargs)
File “xxx/lib/python3.7/site-packages/hug/interface.py”, line 698, in call\_function
return self.interface(\*\*parameters)
File “xxx/lib/python3.7/site-packages/hug/interface.py”, line 100, in **call**
return \_\_hug\_internal\_self.\_function(\*args, \*\*kwargs)
File “xxx/lib/python3.7/site-packages/prodigy/\_api/hug\_app.py”, line 228, in get\_session\_questions
tasks = controller.get\_questions(session\_id=session\_id)
File “cython\_src/prodigy/core.pyx”, line 130, in prodigy.core.Controller.get\_questions
File “cython\_src/prodigy/components/feeds.pyx”, line 58, in prodigy.components.feeds.SharedFeed.get\_questions
File “cython\_src/prodigy/components/feeds.pyx”, line 63, in prodigy.components.feeds.SharedFeed.get\_next\_batch
File “cython\_src/prodigy/components/feeds.pyx”, line 147, in prodigy.components.feeds.SessionFeed.get\_session\_stream
ValueError: Error while validating stream: no first example. This likely means that your stream is empty.
What is wrong here?
My directory contains image samples only as mentioned in cookbook. Also, how to generate stream for images?
Thanks!
---
### Author:  [@ines](https://support.prodi.gy/u/ines)
#### Post date: [June 21, 2019, 2:05pm UTC](https://support.prodi.gy/t/error-while-using-tiny-yolo-on-prodigy-1-8-3/1692/2 "2019-06-21T14:05:30Z")
Hi! The command looks correct 🙂 Based on the error, it seems like for some reason, the model hasn’t sent anything out for annotation and the resulting stream contains no examples. One explanation of this can be that there are just no (valid) image files in the source directory – but that doesn’t seem to be the case here.
It’s also theoretically possible that the model predicted no bounding boxes for any of the images in your samples for any of the [available categories](https://github.com/pjreddie/darknet/blob/master/data/coco.names). It’s unlikely, but possible. Finally, there could also be a bug or some other problem in the experimental Cython implementation of `lightnet` – so you might find that it’s actually much easier to just swap it out for a newer implementation using PyTorch etc. (When we first built this test recipe, there weren’t really any easy-to-use Python options – but that has changed since!)
When you run the recipe with `PRODIGY_LOGGING=basic` enabled, what does the log say? Is it skipping anything, or do you see other statements that look suspicious?