-
Commit 8bd2baf8 has exposed a new failure mode for test_writer.py when run by pypy (only).
nose command: tests/test.py -a !notdefault -x -vv tests/unit/glacier/test_writer.py
nose.config: INFO: I…
-
FairScale FullyShardedDataParallel (FSDP) API supports large model training and is being quickly adopted by internal and external users. The long term goal of upstreaming the API to PyTorch is to rele…
-
## 🐛 Bug
Serialized models have a huge overhead on a first run. With the model I'm using, it takes at least more than 3-4 minutes (than I'm just shutting it down, though it doesn't respond to `SiGS…
-
Hello,
I have been having an issue with erosion not working at all in v2. The same HDA works in v1, but as soon as I switch to v2 it gives the following error. This has been the case with the las…
-
@rekovic @bortigno
Using the standard recipe for L1Ntuples production:
cmsDriver.py -s RAW2DIGI --python_filename=l1NtupleMcMaker2017_RAW2DIGI.py -n 50 --no_output --no_exec --era=Run2_2016 --mc …
-
Svg clipping is not working. Library is rendering image and ignroing clipping
Tested on Xamarin Android with Vapolia.Xamarin.Svg ver 4.1.0, 4.0.11, 4.0.2
Example svg:
```
…
-
`➜ DKRL-master make
g++ Train_cnn_multi.cpp -o Train_cnn_multi -O2 -lpthread
Train_cnn_multi.cpp:296:14: error: expected expression
mut_mutex = PTHREAD_MUTEX_INITIALIZER;
…
-
* 在train()中保存模型
model.save(model_dir)
test()中进行调用时会报错，
model=load_model(model_dir)
---
Traceback (most recent call last):
File ".\MLP.py", line 123, in
test(model_dir,test_generator,te…
d4zzz updated
6 years ago
-
我用于三通道彩图的隐写分析检测，但是我的损失函数一直保持在0.69左右，怎么解决呢
-
Hello ! , im trying to get wich marker has been tracked.
I supposed is with the function "getNyId" ? , but this function is crashing everytime a new marker es read by the cam.