Zhaorun Wu

预测出现全黑问题

5

用voc训练集训练出来的模型，预测时全黑 `# ---------------------------------------------------# # 取出每一个像素点的种类 # ---------------------------------------------------# pr = pr.argmax(axis=-1)` 这个pr是全0的，用预训练模型继续训练，也会出现黑点越来越多的情况，请问这是什么原因呢

How can I get the output of the bboxes

swa_model_12.pth和swa_epoch_12.pth有什么区别吗？

1

还要用get_swa_model()对swa_model_12.pth----swa_model_1.pth求平均吗

遇到了这个错误 gcc: error: unrecognized command line option ‘--showme:version’; did you mean ‘--no-version’?

1

切图时报错，完整信息如下 ``` [i 0908 02:20:53.712638 68 compiler.py:955] Jittor(1.3.5.12) src: /usr/local/miniconda3/envs/jit/lib/python3.8/site-packages/jittor [i 0908 02:20:53.722398 68 compiler.py:956] g++ at /usr/bin/g++(7.5.0) [i 0908 02:20:53.722622 68 compiler.py:957] cache_path: /root/.cache/jittor/jt1.3.5/g++7.5.0/py3.8.13/Linux-4.15.0-1x38/IntelRXeonRSilxdc/default [i 0908 02:20:53.733314 68 __init__.py:411]...

显示forward_ algorithm cache is full和backward w algorithm cache is full对训练有影响吗

1

``` [w 0910 19:24:15.321940 48 cudnn_conv__Tx_float32__Ty_float32__Tw_float32__XFORMAT_abcd__WFORMAT_oihw__YFORMAT_abcd_____hash_4d5b3e2d24c769d3_op.cc:213] forward_ algorithm cache is full [w 0910 19:24:15.391788 84 cudnn_conv__Tx_float32__Ty_float32__Tw_float32__XFORMAT_abcd__WFORMAT_oihw__YFORMAT_abcd_____hash_4d5b3e2d24c769d3_op.cc:213] forward_ algorithm cache is full [w 0910 19:24:15.458474 52 cudnn_conv__Tx_float32__Ty_float32__Tw_float32__XFORMAT_abcd__WFORMAT_oihw__YFORMAT_abcd_____hash_4d5b3e2d24c769d3_op.cc:213] forward_ algorithm cache is...

My validation set is too large. I want to randomly sample 0.1 during validation.

2

The following is my code. ``` val_dataset_sub, _ = random_split(val_dataset, [0.1, 0.9]) val_loader = DataLoader( val_dataset_sub, shuffle=False, batch_size=args.batch_size, num_workers=args.num_workers, ) val_loader = accelerator.prepare(val_loader) model.eval() count = 0 val_losses = {}...

ImportError: No module named 'torchvision'

6

my env is torch 1.4.0 torchvision 0.5.0 cuda 10.1 ``` Traceback (most recent call last): File "D:/Domain-Adaptive-Faster-RCNN-PyTorch/train_net.py", line 18, in from maskrcnn_benchmark.engine.inference import inference File "D:\Domain-Adaptive-Faster-RCNN-PyTorch\maskrcnn_benchmark\engine\inference.py", line 10, in from...