resnet50 fails on inference / not detected by unittests

[undertherain]

[0]blackbird@sugi:benchmarker$ [master|…2]$ python3 -m benchmarker --framework=pytorch --mode=inference --gpus=0 --problem=resnet50 --problem_size=512,3,224,224 --batch_size=32 --precision=FP32 Traceback (most recent call last): File "/usr/lib/python3.8/runpy.py", line 193, in _run_module_as_main return _run_code(code, main_globals, None, File "/usr/lib/python3.8/runpy.py", line 86, in _run_code exec(code, run_globals) File "/home/blackbird/Projects_heavy/performance/benchmarker/benchmarker/main.py", line 14, in main() File "/home/blackbird/Projects_heavy/performance/benchmarker/benchmarker/main.py", line 10, in main run(sys.argv[1:]) File "/home/blackbird/Projects_heavy/performance/benchmarker/benchmarker/benchmarker.py", line 89, in run benchmark.run() File "/home/blackbird/Projects_heavy/performance/benchmarker/benchmarker/modules/i_neural_net.py", line 88, in run results = self.run_internal() File "/home/blackbird/Projects_heavy/performance/benchmarker/benchmarker/modules/do_pytorch.py", line 92, in run_internal x_train, y_train = self.load_data() File "/home/blackbird/Projects_heavy/performance/benchmarker/benchmarker/modules/i_neural_net.py", line 71, in load_data data = get_data(params) File "/home/blackbird/Projects_heavy/performance/benchmarker/benchmarker/util/data/synthetic/img_224_cls_1000.py", line 13, in get_data return gen_classification_data(params, 1000, 224) File "/home/blackbird/Projects_heavy/performance/benchmarker/benchmarker/util/data/synthetic/helpers.py", line 36, in gen_classification_data set_image_size(params, height, width) File "/home/blackbird/Projects_heavy/performance/benchmarker/benchmarker/util/data/synthetic/helpers.py", line 15, in set_image_size params["problem"]["size"] = (params["problem"]["size"],) + shape UnboundLocalError: local variable 'shape' referenced before assignment

[undertherain]

aha, it works if size is not tuple, but one number just need a normal error message