How can i show the pred_score of each class??

[sinyoung-park]

Hi, i'm study CNN to mmclassification in R.O.K. and i study mechainical eng.
i make CNN code in colab follow to mmclassification tutorial.
i make the inference model too.
but the inference model return to pred_label, pred_class, pred_score when i put in image file.
but, i want pred_score of each class.
my dataset is Customdataset.
my dataset num_classse is 5
there are my inference code.

from mmcls.apis import inference_model, init_model
import mmcv

load image

img = mmcv.imread('/content/mmclassification/pcldata/test_set/12slm/12slm1108.png')

initialize model

config_file = '/content/mmclassification/configs/resnet/resnet34_8xb32_in1k.py'
checkpoint_file = '/content/mmclassification/work_dirs/resnet34_8xb32_in1k/latest.pth'
model = init_model(config_file, checkpoint_file, device='cuda')

perform inference

result = inference_model(model, img)
print(result)

how can i fix?? please tell me the sloution..

[Ezra-Yu]

modify this section code https://github.com/open-mmlab/mmclassification/blob/fb16bdc6a25f96843f8fb97b02e45872c61dc566/mmcls/apis/inference.py#L84-L90

To:

    with torch.no_grad():
        scores = model(return_loss=False, **data)
        pred_score = np.max(scores, axis=1)[0]
        pred_label = np.argmax(scores, axis=1)[0]
        result = {'pred_label': pred_label, 'pred_score': float(pred_score), 'scores': scores} # scores is a torch.Tensor. 
    result['pred_class'] = model.CLASSES[result['pred_label']]
    return result

[Ezra-Yu]

new a pred.py file and cp the following code into the file

# Copyright (c) OpenMMLab. All rights reserved.
from argparse import ArgumentParser

import mmcv

from mmcls.apis import my_inference_model, init_model, show_result_pyplot

def my_inference_model(model, img):
    """Inference image(s) with the classifier.
    Args:
        model (nn.Module): The loaded classifier.
        img (str/ndarray): The image filename or loaded image.
    Returns:
        result (dict): The classification results that contains
            `class_name`, `pred_label` and `pred_score`.
    """
    cfg = model.cfg
    device = next(model.parameters()).device  # model device
    # build the data pipeline
    if isinstance(img, str):
        if cfg.data.test.pipeline[0]['type'] != 'LoadImageFromFile':
            cfg.data.test.pipeline.insert(0, dict(type='LoadImageFromFile'))
        data = dict(img_info=dict(filename=img), img_prefix=None)
    else:
        if cfg.data.test.pipeline[0]['type'] == 'LoadImageFromFile':
            cfg.data.test.pipeline.pop(0)
        data = dict(img=img)
    test_pipeline = Compose(cfg.data.test.pipeline)
    data = test_pipeline(data)
    data = collate([data], samples_per_gpu=1)
    if next(model.parameters()).is_cuda:
        # scatter to specified GPU
        data = scatter(data, [device])[0]

    # forward the model
    with torch.no_grad():
        scores = model(return_loss=False, **data)
        pred_score = np.max(scores, axis=1)[0]
        pred_label = np.argmax(scores, axis=1)[0]
        result = {'pred_label': pred_label, 'pred_score': float(pred_score),  'scores': scores}
    result['pred_class'] = model.CLASSES[result['pred_label']]
    return result

def main():
    parser = ArgumentParser()
    parser.add_argument('img', help='Image file')
    parser.add_argument('config', help='Config file')
    parser.add_argument('checkpoint', help='Checkpoint file')
    parser.add_argument(
        '--show',
        action='store_true',
        help='Whether to show the predict results by matplotlib.')
    parser.add_argument(
        '--device', default='cuda:0', help='Device used for inference')
    args = parser.parse_args()

    # build the model from a config file and a checkpoint file
    model = init_model(args.config, args.checkpoint, device=args.device)
    # test a single image
    result = my_inference_model(model, args.img)
    # show the results
    print(mmcv.dump(result, file_format='json', indent=4))
    scores = result['scores']

than run :

python pred.py ${YOUR_IMG_PATH} ${YOUR_CONFIG} ${YOUR_CHECK_POINT}

[Ezra-Yu]

I hope you can understand the code in the script

[sinyoung-park]

i run this model, and i do print(result) return is {'pred_label': 1, 'pred_score': 0.8242934942245483, 'pred_class': '12slm'} it is only one class...

[Ezra-Yu]

It seems you are not running the script I give to you.

Maybe you can try the 1.x branch. refer to this tutorial. which need not modify the code.

[sinyoung-park]

It seems you are not running the script I give to you.

Maybe you can try the 1.x branch. refer to this tutorial. which need not modify the code.

Your code is up and running successfully! Thank you for your help! The problem with this code seems to be a naming error: my_inference_model and inference_model were misnamed. I replaced my_inference_model with infer_image and typed print('Scores',result['scores']) and got the desired result! Thanks for the kind explanation.