aildnont / covid-cxr Goto Github PK

View Code? Open in Web Editor NEW

118.0 118.0 59.0 4.3 MB

Neural network model for classifying chest X-rays by presence of COVID-19 features

License: MIT License

Python 76.72% Jupyter Notebook 23.28%

covid-cxr's People

Contributors

Stargazers

Watchers

Forkers

yamina2019 ayoub-root kamelbelguidoumg2 abd1rahmane mohamedbenz hocinewalid rsingh2083 rafik-ben tzeghnoun erickcai ecode-ethiopia nisargfrd turahul danielbao chaveza9 parag0506 anandkumarh egoisteloim kirtiswagat deepmd-io linhduongtuan quezadasubiabre shankarfierce fashourr eyalho jayshreesuresh yeshvini cicorias ayhkusa pantelis abhisheksingh1979 jmorrison57 adishvakharia jod35 qiaofengxu yabdelsatar subhankar01 iamvazu harendraranasinghe95 gvigkannan gitisara sts-sadr dr-alok-tiwari edmontdants kay360 covid-predictiontool-dtl jeffrey1906 stefandeveloper aaaeeee pragyanaischool jacksky64 summerfang akesar01 dariiakhoroshchuk abc0217 harel-coffee likhithrajanvesh

covid-cxr's Issues

SettingWithCopyWarning

First of all thanks for your effort. When I tried to run the preprocess.py I got the following and the run stop. How can I handle this, please?

preprocess.py:104: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
mila_covid_views_df['label'] = class_dict['COVID-19']
preprocess.py:107: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
mila_views_pneum_df['label'] = class_dict['other_pneumonia'] # Mila CXRs with other peumonias
preprocess.py:110: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
fig1_covid_views_df['label'] = class_dict['COVID-19']

openCV error

while running lime_explain.py I encountered the following error :

Traceback (most recent call last):
  File ".\src\interpretability\lime_explain.py", line 116, in <module>
    explain_xray(lime_dict, i, save_exp=True)                   # Generate explanation for image
  File ".\src\interpretability\lime_explain.py", line 83, in explain_xray
    orig_img = cv2.resize(orig_img, new_dim, interpolation=cv2.INTER_NEAREST)     # Resize image
cv2.error: OpenCV(4.1.0) C:\projects\opencv-python\opencv\modules\imgproc\src\resize.cpp:3718: error: (-215:Assertion failed) !ssize.empty() in function 'cv::resize'

I tried to enclose it into a try-catch block, It skips some images but in subsequent files, the error arises. Like
error: (-209:Sizes of input arguments do not match) The operation is neither 'array op array' (where arrays have the same size and the same number of channels), nor 'array op scalar', nor 'scalar op array' in function 'cv::arithm_op'

FileNotFoundError: [Errno 2] No such file or directory: './data/interpretability/output_class_indices.pkl'

I run train.py file and get this error no output_class_indices.pkl can you help me

lime library issue (unknown opcode) on colab

Hello,
I am running this project on colab. I trained and tested the model on local machine. All stage (training, lime_explanation, batch prediction) are working fine on local machine. On colab training and lime_explanation are working fine. But this error arises during batch prediction stage. The model predicts given image but error arises during explanation stage in predict.py. I have already tried different version of lime in colab.

Found 5 validated image filenames.
Predicting and explaining examples.
100%          1000/1000 [00:06<00:00, 148.94it/s]
---------------------------------------------------------------------------
SystemError                               Traceback (most recent call last)
<ipython-input-1-d9a893551efa> in <module>()
    159 if __name__ == '__main__':
    160     #preds_dir = '/content/final/testImages'
--> 161     results = predict_and_explain_set(preds_dir=None, save_results=True, give_explanations=True)

4 frames
<ipython-input-1-d9a893551efa> in predict_and_explain_set(raw_img_dir, preds_dir, save_results, give_explanations)
    133         # Explain this prediction
    134         if give_explanations:
--> 135             explanation, _ = predict_and_explain(np.squeeze(x, axis=0), model, explainer, NUM_FEATURES, NUM_SAMPLES)
    136             if cfg['LIME']['COVID_ONLY'] == True:
    137                 label_to_see = class_indices['COVID-19']

<ipython-input-1-d9a893551efa> in predict_and_explain(x, model, exp, num_features, num_samples)
     53 
     54     # Generate explanation for the example
---> 55     explanation = exp.explain_instance(x, predict, num_features=num_features, num_samples=num_samples, segmentation_fn=segmentation_fn)
     56     probs = predict_instance(np.expand_dims(x, axis=0), model)
     57     return explanation, probs

/usr/local/lib/python3.6/dist-packages/lime/lime_image.py in explain_instance(self, image, classifier_fn, labels, hide_color, top_labels, num_features, num_samples, batch_size, segmentation_fn, distance_metric, model_regressor, random_seed)
    216                 data, labels, distances, label, num_features,
    217                 model_regressor=model_regressor,
--> 218                 feature_selection=self.feature_selection)
    219         return ret_exp
    220 

/usr/local/lib/python3.6/dist-packages/lime/lime_base.py in explain_instance_with_data(self, neighborhood_data, neighborhood_labels, distances, label, num_features, feature_selection, model_regressor)
    180         """
    181 
--> 182         weights = self.kernel_fn(distances)
    183         labels_column = neighborhood_labels[:, label]
    184         used_features = self.feature_selection(neighborhood_data,

/usr/local/lib/python3.6/dist-packages/lime/lime_image.py in kernel(d, kernel_width)
    118         if kernel is None:
    119             def kernel(d, kernel_width):
--> 120                 return np.sqrt(np.exp(-(d ** 2) / kernel_width ** 2))
    121 
    122         kernel_fn = partial(kernel, kernel_width=kernel_width)

SystemError: unknown opcode

Preprocess.py throws SettingWithCopyWarning

Running preprocess.py results in the following warning:

./src/data/preprocess.py:91: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  mila_other_views_df['label'] = 0                                       # Mila images with alternative diagnoses
./src/data/preprocess.py:92: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  fig1_covid_views_df['label'] = 1                                       # Figure 1 images with COVID-19 diagnosis

getting error in train.py

i am getting this error :
class_ids = list(sorted(class_weight.keys())) AttributeError: 'list' object has no attribute 'keys'

pls help me to fix this

aildnont / covid-cxr Goto Github PK

covid-cxr's People

Contributors

Stargazers

Watchers

Forkers

covid-cxr's Issues

SettingWithCopyWarning

openCV error

FileNotFoundError: [Errno 2] No such file or directory: './data/interpretability/output_class_indices.pkl'

lime library issue (unknown opcode) on colab

Preprocess.py throws SettingWithCopyWarning

getting error in train.py

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent