Analysis of model scripts finds two models

[jeremyfowers]

Example reproduction:

(tkml) jfowers@LAPTOP-5VK03G46:~/onnxmodelzoo/toolchain/models/torchvision/skip$ turnkey ssdlite320_mobilenet_v3_large.py --analyze-only

Models discovered during profiling:

ssdlite320_mobilenet_v3_large.py:
        model (executed 1x - 0.06s)
                Model Type:     Pytorch (torch.nn.Module)
                Class:          SSDLiteFeatureExtractorMobileNet (<class 'torchvision.models.detection.ssdlite.SSDLiteFeatureExtractorMobileNet'>)
                Location:       /home/jfowers/miniconda3/envs/tkml/lib/python3.8/site-packages/torchvision/models/detection/_utils.py, line 461
                Parameters:     3,546,992 (13.53 MB)
                Input Shape:    'Positional Arg 1': (1, 3, 320, 320)
                Hash:           7b1ed851
                Build dir:      /home/jfowers/.cache/turnkey/ssdlite320_mobilenet_v3_large_torchvision_40d8a795

        model (executed 1x - 0.09s)
                Model Type:     Pytorch (torch.nn.Module)
                Class:          SSD (<class 'torchvision.models.detection.ssd.SSD'>)
                Location:       /home/jfowers/miniconda3/envs/tkml/lib/python3.8/site-packages/torchvision/models/detection/ssdlite.py, line 331
                Parameters:     5,198,540 (19.83 MB)
                Input Shape:    'images': (1, 3, 224, 224)
                Hash:           40d8a795
                Build dir:      /home/jfowers/.cache/turnkey/ssdlite320_mobilenet_v3_large_torchvision_40d8a795


Woohoo! The 'benchmark' command is complete.

Desired behavior: Only one model should be discovered.

Full list of model scripts that have this behavior:

• torchvision/ssdlite320_mobilenet_v3_large.py
• rename toolchain/models/timm/{ => skip}/vit_base_r26_s32_224.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_base_r50_s16_224.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_base_r50_s16_224_in21k.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_base_r50_s16_384.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_base_resnet26d_224.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_base_resnet50_224_in21k.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_base_resnet50_384.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_base_resnet50d_224.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_large_r50_s32_224.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_large_r50_s32_224_in21k.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_large_r50_s32_384.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_small_r26_s32_224.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_small_r26_s32_224_in21k.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_small_r26_s32_384.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_small_resnet26d_224.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_small_resnet50d_s16_224.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_tiny_r_s16_p8_224.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_tiny_r_s16_p8_224_in21k.py (100%)
• rename toolchain/models/timm/{ => skip}/vit_tiny_r_s16_p8_384.py (100%)
• rename toolchain/models/torch_hub/{ => skip}/midas_v3_hybrid.py (100%)
• rename toolchain/models/torchvision/{ => skip}/ssd300_vgg16.py (100%)
• rename toolchain/models/transformers/{ => skip}/distil_wav2vec2_for_audio_classification.py (100%)
• rename toolchain/models/transformers/{ => skip}/distilhubert_for_audio_classification.py (100%)
• rename toolchain/models/transformers/{ => skip}/speech_encoder_decoder.py (100%)

cc @danielholanda