KaneAI - World’s First E2E Software Testing Agent.Book a Demo Join TestMu Conference

How to use get_class_names method in lisa

Best Python code snippet using lisa_python

gideon_cloud.py

Source:gideon_cloud.py

...47    print('Layer Output Shape:', layer.output_shape)48    #print('Layer Weights',layer_weights)49print('Total Layers in Network: ', total_layers)50############################### INFERENCE USING KERAS MODEL ###############################51#def get_class_names(path):52def get_class_names():53    class_names = ['abnormal', 'normal']54    #with open(path) as f:55    #    for row in f:56    #        class_names.append(row[:-1])57    #return class_names58#def classify_video(video_path, net_path):59def classify_video(video_path, output_path):60    SAMPLE_DURATION = 16 #Frames Batch Size for input to network for inference61    SAMPLE_SIZE = 11262    #mean = (114.7748, 107.7354, 99.4750)63    #class_names = get_class_names(args.classes)64    class_names = get_class_names()65    # Image/Input Characteristics:66    img_height = 10067    img_width = 10068    channels = 369    batch_size = 1670    # ----------------------------------------- I M P O R T A N T -----------------------------------------71    #winName = 'Deep learning image classification in OpenCV'72    #cv.namedWindow(winName, cv.WINDOW_AUTOSIZE)73        74    # initialize the image mean for mean subtraction along with the75    # predictions queue76    cap = cv.VideoCapture(video_path)77    writer = None78    (W, H) = (None, None)79    # loop over frames from the video file stream80    frames_set = 081    while cv.waitKey(1) < 0:82    # constructing video frames batch for Conv3D input:83        frames = []            84        frames_copy = []85        frames_count = 086            87        for _ in range(SAMPLE_DURATION):88            hasFrame, frame = cap.read()89            if not hasFrame:90                exit(0)91            # clone the output frame, then convert it from BGR to RGB92            output = frame.copy()93            frames_copy.append(output)94                95            # if the frame dimensions are empty, grab them96            if W is None or H is None:97                (H, W) = frame.shape[:2]98                99            frame = cv.resize(frame, (100, 100))100            frame = frame /255101            frames.append(frame)102            frames_count += 1103        frames_set += 1104        print("16-Frames Set# ", frames_set)105        print('Frames Count in processed set: ', frames_count)106            #inputs = cv.dnn.blobFromImages(frames, 1, (SAMPLE_SIZE, SAMPLE_SIZE), mean, True, crop=True)107            #inputs = np.transpose(inputs, (1, 0, 2, 3))108            #inputs = np.expand_dims(inputs, axis=0)109            #net.setInput(inputs)110            #outputs = net.forward()111            #class_pred = np.argmax(outputs)112            #label = class_names[class_pred]113            114            # Performing Prediction on batch of 16 frames:115        frames = np.array(frames).reshape(-1, 16, 100, 100, 3)116        #_ = input('[INFO] >>> Press enter to predict !!!')117        pred_npy = model.predict(frames, batch_size=batch_size, verbose=0)118        #print('Shape of Prediction List: ', pred_npy.shape)119        pred = pred_npy.argmax(axis=1)120        #label = class_names[pred]121            122        threshold = 0.70123        if pred_npy[0][0] >= threshold:124            event = 'Abnormal'125        else:126            event = 'Normal'127        print('Probability (Abnormal):', pred_npy[0][0])128        print('Probability (Normal):', pred_npy[0][1])129        print('Prediction: >> ', event)130            131            # clone the output frame, then convert it from BGR to RGB132            # ordering, resize the frame to a fixed 100x100, and then:133        for frame in frames_copy:134                # draw the activity on the output frame135            text = "Event:  {}".format(event)136            cv.putText(frame, text, (20, 20), cv.FONT_HERSHEY_DUPLEX, 0.50, (0, 255, 0), 2)137            text = "Abnormal Probability: {0:0.2f}".format(pred_npy[0][0])138            cv.putText(frame, text, (20, 40), cv.FONT_HERSHEY_SIMPLEX, 0.50, (0, 255, 0), 2)139            text = "Normal Probability: {0:0.2f}".format(pred_npy[0][1])140            cv.putText(frame, text, (20, 60), cv.FONT_HERSHEY_SIMPLEX, 0.50, (0, 255, 0), 2)141                # check if the video writer is None142            if writer is None:143                    # initialize our video writer144                fourcc = cv.VideoWriter_fourcc(*"H264")145                writer = cv.VideoWriter(output_path, fourcc, 30, (W, H), True)146                # write the output frame to disk147            writer.write(frame)148    print("[INFO] cleaning up system resources...")149    writer.release()150    cap.release()151            #for frame in frames:152            #    labelSize, baseLine = cv.getTextSize(label, cv.FONT_HERSHEY_SIMPLEX, 0.5, 1)153            #    cv.rectangle(frame, (0, 10 - labelSize[1]),154            #                        (labelSize[0], 10 + baseLine), (255, 255, 255), cv.FILLED)155            #    cv.putText(frame, label, (0, 10), cv.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 0))156            #    cv.imshow(winName, frame)157            #if cv.waitKey(1) & 0xFF == ord('q'):158            #    break159            160if __name__ == "__main__":161    args, _ = parser.parse_known_args()162    classify_video(args.input if args.input else 0, args.output)163#def get_class_names(path):164def get_class_names():165    class_names = ['None', 'Abnormal', 'Normal']166    #with open(path) as f:167    #    for row in f:168    #        class_names.append(row[:-1])169    #return class_names170#def classify_video(video_path, net_path):171def classify_video(video_path):172    SAMPLE_DURATION = 16 #Frames Batch Size for input to network for inference173    SAMPLE_SIZE = 112174    #mean = (114.7748, 107.7354, 99.4750)175    #class_names = get_class_names(args.classes)176    class_names = get_class_names()177    # Image/Input Characteristics:178    img_height = 100179    img_width = 100180    channels = 3181    batch_size = 16182    winName = 'Deep learning image classification in OpenCV'183    cv.namedWindow(winName, cv.WINDOW_AUTOSIZE)184    cap = cv.VideoCapture(video_path)185    while cv.waitKey(1) < 0:186        # constructing video frames batch for Conv3D input:187        frames = []188        for _ in range(SAMPLE_DURATION):189            hasFrame, frame = cap.read()190            if not hasFrame:...

debug_list.py

Source:debug_list.py

...11#print(dbio.get_sql_table_as_df('classification_items', addSQL="WHERE classification_id = %s" % 1)["attribute" + str(3)].values)12#validate.aspect_in_db(candidate_file, verbose=True)13#validate.valid_attribute(candidate_file)14#print(validate.create_aspects_table(candidate_file))15#validate.get_class_names(candidate_file)16#validate.check_db_classification("origin_process__1_F_steel_SankeyFlows_2008_Global")17# validate.get_class_names(candidate_file)18# validate.create_db_class_defs(candidate_file)19# validate.create_db_class_items(candidate_file)20# validate.check_classification_definition(validate.get_class_names(candidate_file),21#                                          crash=False, custom_only=False)22# validate.check_classification_items(validate.get_class_names(candidate_file),23#                                     file_io.read_candidate_data_list(candidate_file),24#                                     crash=False, custom_only=False)25# validate.upload_data_list(candidate_file, crash=False)26#dbio.dict_sql_insert(1)27#validate.add_license(candidate_file)28#validate.add_user(candidate_file)29#class_names = validate.get_class_names(file)30#file_data = file_io.read_candidate_data_list(file)31#validate.check_classification_definition(class_names, crash=False)32# validate.create_db_class_defs(file)33# validate.create_db_class_items(file)34#print(validate.check_classification_items(class_names, file_data, crash=False))35#validate.create_db_class_items(file)36exclude_files = [37    '6_MIP_YSTAFDB_MetalUseShares_v1.0.xlsx',  # Done38    '4_PY_YSTAFDB_EoL_RecoveryRate_v1.0.xlsx',  # Done39    # '6_URB_MetabolismOfCities_Jan2019_DOI_7326485.v1.xlsx'  # TODO40    ]41path = IEDC_paths.candidates42focus = []43for file in tqdm(file_io.get_candidate_filenames(path, verbose=1)):44    if file in exclude_files:45        print("Skipping %s" % file)46        continue47    if len(focus) > 0 and file not in focus:48        continue49    try:50        file_meta = file_io.read_candidate_meta(file, path=path)51        aspects_table = validate.create_aspects_table(file_meta)52        class_names = validate.get_class_names(file_meta, aspects_table)53        file_data = file_io.read_candidate_data_list(file, path)54        if not all(validate.check_classification_definition(class_names, crash=False, warn=False)):55            validate.create_db_class_defs(file_meta, aspects_table)56        if not all(validate.check_classification_items(class_names, file_meta, file_data, crash=False, warn=False)):57            validate.create_db_class_items(file_meta, aspects_table, file_data)58        validate.add_user(file_meta, quiet=True)59        validate.add_license(file_meta, quiet=True)60        validate.check_datasets_entry(file_meta, crash_on_exist=False, create=True, update=False, replace=True)61        validate.upload_data_list(file_meta, aspects_table, file_data, crash=False)62    except BaseException as e:63        print("ERROR: File '%s' caused an issue. See stack." % file)...

main.py

Source:main.py

...4from api.core.cnn.cnn_model import CnnModel5app = FastAPI()6cnn = CnnModel()7@app.get("/cnn/class/names")8def get_class_names():9    return cnn.get_class_names()10@app.post("/cnn/classify")11async def classify(file: bytes = File(...)):12    image = Image.open(io.BytesIO(file))13    return cnn.classify(image)14@app.get("/cnn/exists/label/{label}")15def exists_label(label):...

Automation Testing Tutorials

Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.