def classify():
t = time.time()
payload = request.get_json()
if payload is None:
return jsonify({'success': False, 'message': 'must provide sensor data'})
if 'sensor_data' not in payload:
return jsonify({'success': False, 'message': 'must provide sensor data'})
fname = to_base58(payload['sensor_data']['f']) + ".ai"
ai = ai_cache.get(payload['sensor_data']['f'])
if ai is None:
ai = AI(to_base58(payload['sensor_data']['f']))
logger.debug("loading {}".format(fname))
try:
ai.load(fname)
except Exception:
return jsonify({"success": False, "message": "could not find '{p}'".format(p=fname)})
ai_cache[payload['sensor_data']['f']] = ai
classified = ai.classify(payload['sensor_data'])
logger.debug("classifed for {} {:d} ms".format(
payload['sensor_data']['f'], int(1000 * (t - time.time()))))
return jsonify({"success": True, "message": "data analyzed", 'analysis': classified})
def classify():
t = time.time()
payload = request.get_json()
if 'sensor_data' not in payload:
return jsonify({
'success': False,
'message': 'must provide sensor data'
})
data_folder = '.'
if 'data_folder' in payload:
data_folder = payload['data_folder']
fname = os.path.join(data_folder,
to_base58(payload['sensor_data']['f']) + ".find3.ai")
ai = AI(to_base58(payload['sensor_data']['f']), data_folder)
logger.debug("loading {}".format(fname))
try:
ai.load(fname)
except FileNotFoundError:
return jsonify({
"success": False,
"message": "could not find '{p}'".format(p=fname)
})
classified = ai.classify(payload['sensor_data'])
logger.debug("{:d} ms".format(int(1000 * (t - time.time()))))
return jsonify({
"success": True,
"message": "data analyzed",
'analysis': classified
})
def classify(payload):
if payload is None:
return {'success': False, 'message': 'must provide sensor data'}
if 'sensor_data' not in payload:
return {'success': False, 'message': 'must provide sensor data'}
t = time.time()
data_folder = (payload['data_folder'] if 'data_folder' in payload else DEFAULT_DATA_DIRECTORY)
fname = out_file(data_folder, payload['sensor_data']['f'])
ai = ai_cache.get(payload['sensor_data']['f'])
if ai == None:
ai = AI(to_base58(payload['sensor_data']['f']))
logger.debug("loading {}".format(fname))
try:
ai.load(fname, redis_cache=True)
except FileNotFoundError:
logger.error('File not found')
return {"success": False, "message": "could not find '{p}'".format(p=fname)}
ai_cache[payload['sensor_data']['f']] = ai
classified = ai.classify(payload['sensor_data'])
logger.debug("classifed for {} {:d} ms".format(
payload['sensor_data']['f'], int(1000 * (t - time.time()))))
return {"success": True, "message": "data analyzed", 'analysis': classified}
def learn():
payload = request.get_json()
if 'family' not in payload:
return jsonify({'success': False, 'message': 'must provide family'})
if 'csv_file' not in payload:
return jsonify({'success': False, 'message': 'must provide CSV file'})
data_folder = '.'
if 'data_folder' in payload:
data_folder = payload['data_folder']
else:
logger.debug("could not find data_folder in payload")
logger.debug(data_folder)
ai = AI(to_base58(payload['family']), data_folder)
fname = os.path.join(data_folder, payload['csv_file'])
try:
ai.learn(fname)
except FileNotFoundError:
return jsonify({
"success": False,
"message": "could not find '{}'".format(fname)
})
ai.save(
os.path.join(data_folder,
to_base58(payload['family']) + ".find3.ai"))
return jsonify({"success": True, "message": "calibrated data"})
def learn(payload):
if payload is None:
return {'success': False, 'message': 'must provide sensor data'}
if 'family' not in payload:
return {'success': False, 'message': 'must provide family'}
if 'csv_file' not in payload:
return {'success': False, 'message': 'must provide CSV file'}
data_folder = '.'
if 'data_folder' in payload:
data_folder = payload['data_folder']
else:
logger.debug("could not find data_folder in payload")
logger.debug(data_folder)
ai = AI(to_base58(payload['family']), data_folder)
fname = os.path.join(data_folder, payload['csv_file'])
try:
ai.learn(fname)
except FileNotFoundError:
return {
"success": False,
"message": "could not find '{}'".format(fname)
}
print(payload['family'])
ai.save(
os.path.join(data_folder,
to_base58(payload['family']) + ".find3.ai"))
ai_cache[payload['family']] = ai
return {"success": True, "message": "calibrated data"}
def classify(payload):
if payload is None:
return {'success': False, 'message': 'must provide sensor data'}
if 'sensor_data' not in payload:
return {'success': False, 'message': 'must provide sensor data'}
t = time.time()
data_folder = '.'
if 'data_folder' in payload:
data_folder = payload['data_folder']
fname = os.path.join(data_folder,
to_base58(payload['sensor_data']['f']) + ".find3.ai")
ai = ai_cache.get(payload['sensor_data']['f'])
if ai == None:
ai = AI(to_base58(payload['sensor_data']['f']), data_folder)
logger.debug("loading {}".format(fname))
try:
ai.load(fname)
except FileNotFoundError:
logger.error('File not found')
return {
"success": False,
"message": "could not find '{p}'".format(p=fname)
}
ai_cache[payload['sensor_data']['f']] = ai
classified = ai.classify(payload['sensor_data'])
logger.debug("classifed for {} {:d} ms".format(
payload['sensor_data']['f'], int(1000 * (t - time.time()))))
return {
"success": True,
"message": "data analyzed",
'analysis': classified
}
def learn(payload):
if payload is None:
return {'success': False, 'message': 'must provide sensor data'}
if 'family' not in payload:
return {'success': False, 'message': 'must provide family'}
if 'csv_file' not in payload and 'file_data' not in payload:
return {'success': False, 'message': 'must provide CSV file'}
data_folder = (payload['data_folder'] if 'data_folder' in payload else path_to_data)
ai = AI(to_base58(payload['family']))
# encoded file in request payload
if 'file_data' in payload:
ai.learn("", file_data=payload['file_data'])
# # file on disk
# # requires absolute path
# elif 'csv_file' in payload:
# try:
# ai.learn( payload['csv_file'] )
# except FileNotFoundError:
# return {"success": False, "message": "could not find '{0}'".format( payload['csv_file'] )}
fname = out_file(data_folder, payload['family'])
ai.save(fname, redis_cache=True)
ai_cache[payload['family']] = ai
return {"success": True, "message": "calibrated data"}
def learn():
payload = request.get_json()
family = 'posifi'
if payload is None:
return jsonify({'success': False, 'message': 'must provide sensor data'})
if 'csv_file' not in payload:
return jsonify({'success': False, 'message': 'must provide CSV file'})
ai = AI(to_base58(family))
try:
ai.learn(payload['csv_file'])
except Exception as e:
return jsonify({"success": False, "message": f"ERROR learning {e}"})
ai.save(to_base58(family) + ".ai")
ai_cache[family] = ai
return jsonify({"success": True, "message": "calibrated data"})