Can you please explain to me the following Type Error. I am getting this error when I try to upload the test image via a webpage. But it works fine when I CURL it
2021-03-31 21:54:09,364 [INFO ] W-9000-densenet161_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - image = torch.FloatTensor(image)
**21-03-31 21:54:09,364 [INFO ] W-9000-densenet161_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - TypeError: new(): data must be a sequence (got NoneType)
Based on the error message image is not loaded properly and is a None as seen here:
torch.FloatTensor(None)
> TypeError: new(): data must be a sequence (got NoneType)
so I assume the upload or data loading fails.
@ptrblck thank you so much for having a look. Please see below the logs. Can you please let me know what can be the possible reason for " Invoking custom service failed"
2021-04-01 12:40:27,029 [INFO ] W-9001-densenet161_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - Invoking custom service failed.
2021-04-01 12:40:27,030 [INFO ] W-9001-densenet161_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - Traceback (most recent call last):
2021-04-01 12:40:27,031 [INFO ] W-9001-densenet161_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - File “/home/venv/lib/python3.6/site-packages/ts/service.py”, line 100, in predict
2021-04-01 12:40:27,032 [INFO ] W-9001-densenet161_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - ret = self._entry_point(input_batch, self.context)
2021-04-01 12:40:27,032 [INFO ] W-9001-densenet161_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - File “/home/venv/lib/python3.6/site-packages/ts/torch_handler/base_handler.py”, line 194, in handle
2021-04-01 12:40:27,033 [INFO ] W-9001-densenet161_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - data_preprocess = self.preprocess(data)
2021-04-01 12:40:27,034 [INFO ] W-9001-densenet161_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - File “/home/venv/lib/python3.6/site-packages/ts/torch_handler/vision_handler.py”, line 49, in preprocess
2021-04-01 12:40:27,039 [INFO ] W-9001-densenet161_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - image = torch.FloatTensor(image)
2021-04-01 12:40:27,040 [INFO ] W-9001-densenet161_1.0-stdout org.pytorch.serve.wlm.WorkerLifeCycle - TypeError: new(): data must be a sequence (got NoneType)
2021-04-01 12:40:27,032 [INFO ] W-9001-densenet161_1.0 org.pytorch.serve.wlm.WorkerThread - Backend response time: 2
2021-04-01 12:40:27,041 [INFO ] W-9001-densenet161_1.0 ACCESS_LOG - /192.168.8.115:35886 “POST /predictions/densenet161 HTTP/1.1” 503 123
2021-04-01 12:40:27,041 [INFO ] W-9001-densenet161_1.0 TS_METRICS - Requests5XX.Count:1|#Level:Host|#hostname:6b80040fb46b,timestamp:null
2021-04-01 12:40:27,050 [DEBUG] W-9001-densenet161_1.0 org.pytorch.serve.job.Job - Waiting time ns: 101351455, Inference time ns: 124289758
2021-04-01 12:40:27,051 [INFO ] W-9001-densenet161_1.0 TS_METRICS - WorkerThreadTime.ms:22|#Level:Host|#hostname:6b80040fb46b,timestamp:null
This error message is most likely raised by the new() error, as it’s also pointing to it in the stack trace a bit later.