Feature/ml 356 ml synergetic contract

.gitignore

@@ -1 +1,2 @@
 .DS_Store
+.idea/

ml_examples/01_smart_contract_model/model.etch

@@ -0,0 +1,111 @@
+//------------------------------------------------------------------------------
+//
+//   Copyright 2019 Fetch.AI Limited
+//
+//   Licensed under the Apache License, Version 2.0 (the "License");
+//   you may not use this file except in compliance with the License.
+//   You may obtain a copy of the License at
+//
+//       http://www.apache.org/licenses/LICENSE-2.0
+//
+//   Unless required by applicable law or agreed to in writing, software
+//   distributed under the License is distributed on an "AS IS" BASIS,
+//   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+//   See the License for the specific language governing permissions and
+//   limitations under the License.
+//
+//------------------------------------------------------------------------------
+
+persistent model_state : Model;
+persistent data_state : Tensor;
+persistent label_state : Tensor;
+
+// initial set up creates the model and persistent data
+@init
+function setup(owner : Address)
+  use model_state;
+  use data_state;
+  use label_state;
+
+  // set up the initial model
+  var model = model_state.get(Model("sequential"));
+  model.add("dense", 13u64, 10u64, "relu");
+  model.add("dense", 10u64, 10u64, "relu");
+  model.add("dense", 10u64, 1u64);
+  model.compile("mse", "adam");
+  model_state.set(model);
+
+  // define the initial input data
+  var data_shape = Array<UInt64>(2);
+  data_shape[0] = 13u64;                // boston feature size is 13
+  data_shape[1] = 1u64;                 // batch size == 1
+  var data = data_state.get(Tensor(data_shape));
+  data_state.set(data);
+
+  // define the initial label
+  var label_shape = Array<UInt64>(2);
+  label_shape[0] = 1u64;                // house price output size is 1
+  label_shape[1] = 1u64;                // batch size == 1
+  var label = label_state.get(Tensor(label_shape));
+  label_state.set(label);
+
+endfunction
+
+// get the data state
+@query
+function getData() : String
+  use data_state;
+  var data = data_state.get();
+  return data.toString();
+endfunction
+
+// get the label state
+@query
+function getLabel() : String
+  use label_state;
+  var label = label_state.get();
+  return label.toString();
+endfunction
+
+// get the current training loss of the model
+@query
+function evaluate() : String
+  use model_state;
+  var model = model_state.get();
+  var loss = model.evaluate();
+  var str_loss = toString(loss[0]);
+  return str_loss;
+endfunction
+
+// pass in some data, train the model with it, save the updated model to state
+@action
+function train(data_string: String, label_string: String)
+  use model_state;
+  var model = model_state.get();
+
+  use data_state;
+  var data = data_state.get();
+  data.fromString(data_string);
+
+  use label_state;
+  var label = label_state.get();
+  label.fromString(label_string);
+
+  var batch_size = 10u64;
+  model.fit(data, label, batch_size);
+  model_state.set(model);
+endfunction
+
+// make a prediction with the model based on input data passed to function
+@query
+function predict(data_string: String) : String
+  use model_state;
+  var model = model_state.get();
+
+  use data_state;
+  var data = data_state.get();
+  data.fromString(data_string);
+
+  var prediction = model.predict(data);
+  return prediction.toString();
+endfunction

ml_examples/01_smart_contract_model/submit_contract.py

@@ -0,0 +1,67 @@
+from fetchai.ledger.api import LedgerApi
+from fetchai.ledger.contract import Contract
+from fetchai.ledger.crypto import Entity, Address
+import sys
+import time
+
+DATA_FILE = "/Users/khan/fetch/corpora/boston/boston_data.csv"
+LABEL_FILE = "/Users/khan/fetch/corpora/boston/boston_label.csv"
+
+def read_one_line_data_csv_as_string(fname):
+
+    f = open(fname, 'r')
+    return f.readline()
+
+
+def main(source, train_data, train_labels, test_data, test_labels):
+
+    # Create keypair for the contract owner
+    entity = Entity()
+    address = Address(entity)
+
+    # Setting API up
+    api = LedgerApi('127.0.0.1', 8000)
+
+    # Need funds to deploy contract
+    api.sync(api.tokens.wealth(entity, 10000000000000))
+
+    # Create contract
+    contract = Contract(source, entity)
+
+    # Deploy contract
+    # api.sync(api.contracts.create(entity, contract, 1000000000))
+    api.sync(contract.create(api, entity, 1000000000))
+
+    # grab the data and label tensor - just to demonstrate usage
+    data_string = contract.query(api, 'getData')
+    label_string = contract.query(api, 'getLabel')
+
+    # grab boston data set data
+    data_string = read_one_line_data_csv_as_string(DATA_FILE)
+    label_string = read_one_line_data_csv_as_string(LABEL_FILE)
+
+    print("initial data: " + data_string)
+    print("initial label: " + label_string)
+
+    # train on some input data
+    fet_tx_fee = 16000000
+    api.sync(contract.action(api, 'train', fet_tx_fee, [entity], data_string, label_string))
+
+    # evaluate the initial loss
+    initial_loss = contract.query(api, 'evaluate')
+
+    # predict loss after training
+    prediction = contract.query(api, 'predict', data_string=data_string)
+    print("model prediction: " + prediction)
+
+if __name__ == '__main__':
+
+    # Loading contract
+    if len(sys.argv) != 6:
+      print("Usage: ", sys.argv[0], "[filename] train_data.csv train_labels.csv test_data.csv test_labels.csv")
+      exit(-1)
+
+    with open(sys.argv[1], "r") as fb:
+      source = fb.read()
+
+    main(source, sys.argv[2], sys.argv[3], sys.argv[4], sys.argv[5])

ml_examples/02_synergetic_contract_model/model.etch

@@ -0,0 +1,141 @@
+//------------------------------------------------------------------------------
+//
+//   Copyright 2019 Fetch.AI Limited
+//
+//   Licensed under the Apache License, Version 2.0 (the "License");
+//   you may not use this file except in compliance with the License.
+//   You may obtain a copy of the License at
+//
+//       http://www.apache.org/licenses/LICENSE-2.0
+//
+//   Unless required by applicable law or agreed to in writing, software
+//   distributed under the License is distributed on an "AS IS" BASIS,
+//   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+//   See the License for the specific language governing permissions and
+//   limitations under the License.
+//
+//------------------------------------------------------------------------------
+
+persistent model_state : Model;
+persistent data_state : Tensor;
+persistent label_state : Tensor;
+persistent loss_state : Fixed64;
+
+// TODO : createProblem must return the problem - this should wrap up model and data somehow
+// TODO : doWork must take in the problem & nonce, and return the solution - solution must be the new model
+// TODO : evaluateWork must take in the solution, and return a value
+// TODO : applyWork will save the new model to a state, for querying
+
+@init
+function setup()
+  use model_state;
+  use data_state;
+  use label_state;
+
+  // set up the initial model
+  var model = model_state.get(Model("sequential"));
+  model.add("dense", 13u64, 10u64, "relu");
+  model.add("dense", 10u64, 10u64, "relu");
+  model.add("dense", 10u64, 1u64);
+  model.compile("mse", "adam");
+  model_state.set(model);
+
+  // define the initial input data
+  var data_shape = Array<UInt64>(2);
+  data_shape[0] = 13u64;                // boston feature size is 13
+  data_shape[1] = 1u64;                 // batch size == 1
+  var data = data_state.get(Tensor(data_shape));
+  data_state.set(data);
+
+  // define the initial label
+  var label_shape = Array<UInt64>(2);
+  label_shape[0] = 1u64;                // house price output size is 1
+  label_shape[1] = 1u64;                // batch size == 1
+  var label = label_state.get(Tensor(label_shape));
+  label_state.set(label);
+
+endfunction
+
+// set up a problem around training a machine learning model
+@problem
+function createProblem(model: Model, data: Tensor, label: Tensor)
+
+  return {model, data, label}
+
+endfunction
+
+// get the data state
+@query
+function getData() : String
+  use data_state;
+  var data = data_state.get();
+  return data.toString();
+endfunction
+
+// get the label state
+@query
+function getLabel() : String
+  use label_state;
+  var label = label_state.get();
+  return label.toString();
+endfunction
+
+// evaluates performance as the loss function of the model after training
+@objective
+function evaluateWork(in_model: String)
+  use model_state;
+  var model = model_state.get();
+  model.fromString(in_model);
+  return model.evaluate();
+endfunction
+
+// the work of training the model
+@objective
+function doWork(problem : String, nonce : UInt256) :  String
+  use model_state;
+  use data_state;
+  use label_state;
+
+  var model = model_state.get();
+  model.fromString(in_model);
+
+  // update the learning rate of the local model
+  var lr = nonce.toFloat64() % 1.0fp64;
+  model.setLearningRate(lr);
+
+  // train the model
+  var batch_size = 10u64;
+  var data = data_state.get();
+  var label = label_state.get();
+  model.fit(data, label, batch_size);
+
+  // return the serialised model
+  return model.toString();
+endfunction
+
+// set the new model to be the specified winner
+@clear
+function applyWork(in_model: String)
+  use model_state;
+  var model = model_state.get();
+  model.fromString(in_model);
+  model_state.set(model);
+endfunction
+
+// make a prediction with the model based on input data passed to function
+@query
+function predict(data: Tensor) : Tensor
+  use model_state;
+  var model = model_state.get();
+  var prediction = model.predict(data);
+  return prediction;
+endfunction
+
+// query the current model performance
+@query
+function evaluate() : Tensor
+  use model_state;
+  var model = model_state.get();
+  return model.evaluate();
+endfunction
+

ml_examples/02_synergetic_contract_model/submit_contract.py

@@ -0,0 +1,72 @@
+from fetchai.ledger.api import LedgerApi
+from fetchai.ledger.contract import Contract
+from fetchai.ledger.crypto import Entity, Address
+import sys
+import time
+
+
+def main(source, train_data, train_labels, test_data, test_labels):
+
+
+    # Create keypair for the contract owner
+    entity = Entity()
+    address = Address(entity)
+
+    # Setting API up
+    api = LedgerApi('127.0.0.1', 8000)
+
+    # Need funds to deploy contract
+    api.sync(api.tokens.wealth(entity, 10000000000000))
+
+    # Create contract
+    contract = Contract(source, entity)
+
+    # Deploy contract
+    api.sync(contract.create(api, entity, 1000000000))
+
+    # Run the synergetic contract
+
+    # combine the data and label into one string for passing into the createProblem function
+    data_string = contract.query(api, 'getData')
+    label_string = contract.query(api, 'getLabel')
+    print("initial data: " + data_string)
+    print("initial label: " + label_string)
+
+    # TODO : combine data and label
+
+    # pass the combined input data
+    api.sync(api.contracts.submit_data(entity, contract.digest, contract.address, value = (data_string + label_string)))
+    api.wait_for_blocks(10)
+    result = contract.query(api, 'query_result')
+
+
+
+    # evaluate the initial loss
+    initial_loss = contract.query(api, 'evaluate')
+    print("initial loss: " + initial_loss)
+
+    # grab the data and label tensor
+    data_string = contract.query(api, 'getData')
+    label_string = contract.query(api, 'getLabel')
+    print("initial data: " + data_string)
+    print("initial label: " + label_string)
+
+    # train on some input data
+    fet_tx_fee = 16000000
+    api.sync(contract.action(api, 'train', fet_tx_fee, [entity], data_string, label_string))
+
+    # predict loss after training
+    prediction = contract.query(api, 'predict', data_string=data_string)
+    print("model prediction: " + prediction)
+
+if __name__ == '__main__':
+
+    # Loading contract
+    if len(sys.argv) != 6:
+      print("Usage: ", sys.argv[0], "[filename] train_data.csv train_labels.csv test_data.csv test_labels.csv")
+      exit(-1)
+
+    with open(sys.argv[1], "r") as fb:
+      source = fb.read()
+
+    main(source, sys.argv[2], sys.argv[3], sys.argv[4], sys.argv[5])