lint(api): remove some confusing log messages

api/onnx_web/convert/correction/gfpgan.py

@@ -27,7 +27,6 @@ def convert_correction_gfpgan(
         logger.info("ONNX model already exists, skipping")
         return
 
-    logger.info("loading and training model")
     model = RRDBNet(
         num_in_ch=3,
         num_out_ch=3,
@@ -38,7 +37,6 @@ def convert_correction_gfpgan(
     )
 
     torch_model = torch.load(source, map_location=conversion.map_location)
-    # TODO: make sure strict=False is safe here
     if "params_ema" in torch_model:
         model.load_state_dict(torch_model["params_ema"], strict=False)
     else:

api/onnx_web/convert/upscaling/bsrgan.py

@@ -27,7 +27,6 @@ def convert_upscaling_bsrgan(
         logger.info("ONNX model already exists, skipping")
         return
 
-    logger.info("loading and training model")
     # values based on https://github.com/cszn/BSRGAN/blob/main/main_test_bsrgan.py#L69
     model = RRDBNet(
         num_in_ch=3,

api/onnx_web/convert/upscaling/resrgan.py

@@ -31,8 +31,6 @@ def convert_upscale_resrgan(
         logger.info("ONNX model already exists, skipping")
         return
 
-    logger.info("loading and training model")
-
     if TAG_X4_V3 in name:
         # the x4-v3 model needs a different network
         model = SRVGGNetCompact(

api/onnx_web/convert/upscaling/swinir.py

@@ -27,7 +27,6 @@ def convert_upscaling_swinir(
         logger.info("ONNX model already exists, skipping")
         return
 
-    logger.info("loading and training model")
     # values based on https://github.com/JingyunLiang/SwinIR/blob/main/main_test_swinir.py#L128
     params = {
         "depths": [6] * 6,