def predict(context, exported_model):
"""Start a repl executing examples against a caffe2 model."""
config = context.obj.load_config()
print(f"Loading model from {exported_model or config.export_caffe2_path}")
predictor = create_predictor(config, exported_model)
print(f"Model loaded, reading example JSON from stdin")
for line in sys.stdin.readlines():
input = json.loads(line)
predictions = predictor(input)
pprint.pprint(predictions)
import sys
import flask
import pytext
config_file = sys.argv[1]
model_file = sys.argv[2]
config = pytext.load_config(config_file)
predictor = pytext.create_predictor(config, model_file)
app = flask.Flask(__name__)
@app.route('/get_flight_info', methods=['GET', 'POST'])
def get_flight_info():
text = flask.request.data.decode()
# Pass the inputs to PyText's prediction API
result = predictor({"raw_text": text})
# Results is a list of output blob names and their scores.
# The blob names are different for joint models vs doc models
# Since this tutorial is for both, let's check which one we should look at.
doc_label_scores_prefix = (
'scores:' if any(r.startswith('scores:') for r in result)
else 'doc_scores:'
)
# For now let's just output the top document label!
best_doc_label = max(
(label for label in result if label.startswith(doc_label_scores_prefix)),
key=lambda label: result[label][0],
import sys
import pytext
config_file = sys.argv[1]
model_file = sys.argv[2]
config = pytext.load_config(config_file)
predictor = pytext.create_predictor(config, model_file)
text = input('\nPlease Enter the text\n')
result = predictor({"raw_text": text})
doc_label_scores_prefix = (
'scores:' if any(r.startswith('scores:') for r in result)
else 'doc_scores:'
)
# For now let's just output the top document label!
best_doc_label = max(
(label for label in result if label.startswith(doc_label_scores_prefix)),
key=lambda label: result[label][0],
# Strip the doc label prefix here
)[len(doc_label_scores_prefix):]
<<<<<<< HEAD
print("-> ",best_doc_label)
=======
print(best_doc_label)
>>>>>>> ef352d25559a45f84c010790e473a30fc6ce5526
word_scores = {label.split(':')[1]: result[label] for label in result if label.startswith(slot_prefix)}
word_labels = []
for i in range(len(word_scores[list(word_scores.keys())[0]])):
best_label = '--'
best_score = 100
for label in word_scores.keys():
if word_scores[label][i] < best_score:
best_score = word_scores[label][i]
best_label = label
word_labels.append(best_label)
return word_labels
#doc_label_scores_prefix = ('scores:' if any(r.startswith('scores:') for r in result) else 'doc_scores:')
#return max((label for label in result if label.startswith(doc_label_scores_prefix)), key=lambda label: result[label][0],)[len(doc_label_scores_prefix):]
configp = pytext.load_config(config['paths']['etc']['pytext']['model-config-extended'])
predictor = pytext.create_predictor(configp, config['paths']['etc']['pytext']['model-extended'])
test_dataset = fo.read_json(config['paths']['datasets']['pytext']['test-extended'])
counter = 0
positive_counter = 0
total_recall = []
#print(f"{'Sample':80s}\t{'recognized-label':20s}\t{'true-label':20s}\t{'correctly-recognized':30s}")
for label in test_dataset.keys():
for sample in test_dataset[label]:
recognized = [slot for slot in get_best_slots(predictor({"text": sample["text"].lower(), "doc_weight": 1, "word_weight": 1}))]# if slot != '__UNKNOWN__']
parsed_command = list(zip(sample['text'].lower().split(' '), recognized))
parsed_right_command = list(zip(sample['text'].lower().split(' '), sample['slots']))
print(f"-- Recognized slots")
ul.get_request_type(parsed_command, filename = config['paths']['datasets']['request-mapping']['requests'])
print(f"-- Right slots")
