def predict(self, data):
# TODO: this evaluates a single example, i.e. mini-batch of one.
# generalize this to general case.
if len(data.shape) == 1:
data = data.reshape((1, data.shape[0]))
classification_response = self.skil.api.multipredict(
deployment_name=self.deployment.name,
model_name=self.model_name,
version_name="default",
body=skil_client.MultiPredictRequest(
id=str(uuid.uuid1()),
needs_pre_processing=False,
inputs=[
skil_client.INDArray(ordering='c',
shape=list(data.shape),
data=data.tolist()[0]),
skil_client.
INDArray( # This is the keep_prob placeholder data
ordering='c',
shape=[1],
data=[1.0])
]))
output = classification_response.outputs[0]
prediction = np.asarray(output.data)
shape = output.shape
return prediction.reshape(shape)
def predict_single(self, data, version='default'):
"""Predict for a single input.
# Arguments:
data: `numpy.ndarray` (or list thereof). Input data.
version: version of the deployed service
# Returns
`numpy.ndarray` instance for single output model and list of `numpy.ndarray` for multi-output model.
"""
if isinstance(data, list):
inputs = [self._indarray(np.expand_dims(x, 0)) for x in data]
else:
inputs = [self._indarray(np.expand_dims(data, 0))]
classification_response = self.skil.api.multipredict(
deployment_name=self.deployment.name,
model_name=self.model_name,
version_name=version,
body=skil_client.MultiPredictRequest(
id=str(uuid.uuid1()),
needs_pre_processing=False,
inputs=inputs
)
)
output = classification_response.outputs[0]
return np.asarray(output.data).reshape(output.shape)
def predict(self, data, version='default'):
"""Predict for given batch of data.
# Arguments:
data: `numpy.ndarray` (or list thereof). Batch of input data, or list of batches for multi-input model.
version: version of the deployed service
# Returns
`numpy.ndarray` instance for single output model and list of `numpy.ndarray` for multi-output model.
"""
if isinstance(data, list):
inputs = [self._indarray(x) for x in data]
else:
inputs = [self._indarray(data)]
classification_response = self.skil.api.multipredict(
deployment_name=self.deployment.name,
model_name=self.model_name,
version_name=version,
body=skil_client.MultiPredictRequest(
id=str(uuid.uuid1()),
needs_pre_processing=False,
inputs=inputs
)
)
outputs = classification_response.outputs
outputs = [np.asarray(o.data).reshape(o.shape) for o in outputs]
if len(outputs) == 1:
return outputs[0]
return outputs
r = requests.post("http://192.168.1.128:9008/login", json={"userId": "admin", "password": "******"})
token = r.json()['token']
configuration.api_key['authorization'] = f'Bearer {token}'
api_instance = skil_client.DefaultApi(skil_client.ApiClient(configuration))
list_ind_array = [[convert_indarray(np.expand_dims(image_batch[i,:,:,:], axis=0))] for i in range(12)]
batch_results = []
index = 0
for data in list_ind_array:
print("getting response for batch image ", index)
body_data = skil_client.MultiPredictRequest(
id=str(uuid.uuid1()),
needs_pre_processing=False,
inputs=data
)
response = api_instance.multipredict("age", "default", "outputgraphwithsoftmax", body=body_data )
response = response.to_dict()
output = response['outputs'][0]
probabilities = output['data']
probabilities = np.array(probabilities)
batch_results.append(probabilities)
index+=1
time.sleep(0.5) #prevent spamming
output = batch_results[0]
batch_sz = len(batch_results)
arg_max_each_batch = []