有没有办法将多个图像发送到 tf 服务 model？

Question

I'm trying this below code but I got unexpected error

This is my code for getting input and pass it to model.

def get_instances(dir = '/test_data'):

        instances = list()
        file_names = [file.split('/')[-1] for file in os.listdir(dir)]
        for file in file_names :
          image = nv.imread(os.path.join(dir ,file), resize = (300,300), color_mode='rgb',normalize=True)
          image = combine_rgb_xyz(image)
          #image = nv.expand_dims(image,axis=0)
          instances.append(image)
        return np.array(instances) ,file_names 

After I send these data to model with below code:

def make_prediction(instances):
   url = get_url()
   data = json.dumps({"signature_name": "serving_default", "instances": instances.tolist()})
   headers = {"content-type": "application/json"}
   json_response = requests.post(url, data=data, headers=headers)
   predictions = json.loads(json_response.text)['predictons']
   return predictions

but I get unexpected output:

'predictons'

Answer 1

You can use tf.make_tensor_proto and tf.make_ndarray for image numpy array to/from tensor conversions. Then, you can use 'serving_default' signature to make predictions and pass multiple images to serving_default request to achieve faster results. 'serving_default' signature supports multiple images to be processed at once since it has 4d input (batch, height, width, channel). Please refer this guide to pass multiple images to TF serving.