Merge pull request #61 from Microsoft/master

merge master

README.md

@@ -37,7 +37,7 @@ Note: If you are in docker container (as root), please remove `--user` from the
 * We only support Linux (Ubuntu 16.04 or higher) in our current stage. 
 * Run the following commands in an environment that has `python >= 3.5`, `git` and `wget`.
 ```bash	
-    git clone -b v0.3.3 https://github.com/Microsoft/nni.git	
+    git clone -b v0.3.4 https://github.com/Microsoft/nni.git
     cd nni	
     source install.sh	
 ```
@@ -47,7 +47,7 @@ Note: If you are in docker container (as root), please remove `--user` from the
 The following example is an experiment built on TensorFlow. Make sure you have **TensorFlow installed** before running it.	
 * Download the examples via clone the source code.	
 ```bash	
-    git clone -b v0.3.3 https://github.com/Microsoft/nni.git
+    git clone -b v0.3.4 https://github.com/Microsoft/nni.git
 ```
 * Run the mnist example.
 ```bash

deployment/docker/Dockerfile

@@ -18,7 +18,7 @@
 # DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 
-FROM nvidia/cuda:9.0-cudnn7-devel-ubuntu16.04
+FROM nvidia/cuda:9.0-cudnn7-runtime-ubuntu16.04
 
 LABEL maintainer='Microsoft NNI Team<[email protected]>'
 
@@ -53,7 +53,7 @@ RUN python3 -m pip --no-cache-dir install \
 #
 #Install NNI
 #
-RUN python3 -m pip install --user nni
+RUN python3 -m pip install nni
 
 #
 #Tensorflow 1.10.0
@@ -68,6 +68,6 @@ RUN python3 -m pip --no-cache-dir install Keras==2.1.6
 #sklearn
 RUN python3 -m pip --no-cache-dir install scikit-learn
 
-ENV PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/root/.local/bin:/usr/bin:
+ENV PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/root/.local/bin:/usr/bin:/bin:/sbin
 
 WORKDIR /root

deployment/pypi/setup.py

@@ -11,7 +11,7 @@
 
 setuptools.setup(
     name = 'nni',
-    version = '0.3.2',
+    version = '0.3.4',
     author = 'Microsoft NNI team',
     author_email = '[email protected]',
     description = 'Neural Network Intelligence package',
@@ -32,7 +32,8 @@
         'pyyaml',
         'psutil',
         'requests',
-        'astor'
+        'astor',
+        'pyhdfs'
     ],
     classifiers = [
         'Programming Language :: Python :: 3',

docs/CONTRIBUTING.md

@@ -28,7 +28,7 @@ When raising issues, please specify the following:
 
 Provide PRs with appropriate tags for bug fixes or enhancements to the source code. Do follow the correct naming conventions and code styles when you work on and do try to implement all code reviews along the way.
 
-If you are looking for How to go about contributing and debugging the NNI source code, you can refer our [How to Contribute](./HowToContribute.md) file in the `docs` folder.
+If you are looking for How to develop and debug the NNI source code, you can refer to [How to set up NNI developer environment doc](./SetupNNIDeveloperEnvironment.md) file in the `docs` folder.
 
 Similarly for [writing trials](./WriteYourTrial.md) or [starting experiments](StartExperiment.md). For everything else, refer [here](https://github.com/Microsoft/nni/tree/master/docs).
 

docs/GetStarted.md

@@ -18,7 +18,7 @@
 
 * __Install NNI through source code__
 
-      git clone -b v0.3.3 https://github.com/Microsoft/nni.git
+      git clone -b v0.3.4 https://github.com/Microsoft/nni.git
       cd nni
       source install.sh
 

docs/InstallNNI_Ubuntu.md

@@ -16,7 +16,7 @@
 
 * __Install NNI through source code__
 
-      git clone -b v0.3.3 https://github.com/Microsoft/nni.git
+      git clone -b v0.3.4 https://github.com/Microsoft/nni.git
       cd nni
       source install.sh
 

docs/NNICTLDOC.md

@@ -15,6 +15,7 @@ nnictl experiment
 nnictl config
 nnictl log
 nnictl webui
+nnictl tensorboard
 ```
 ### Manage an experiment
 * __nnictl create__ 
@@ -122,6 +123,23 @@ nnictl webui
           | ------ | ------ | ------ |------ |
           | id|  False| |ID of the experiment you want to set|
           | --value, -v|  True| |the experiment duration will be NUMBER seconds. SUFFIX may be 's' for seconds (the default), 'm' for minutes, 'h' for hours or 'd' for days.|  
+
+ 	* __nnictl update trialnum__  
+        * Description
+
+		      You can use this command to update an experiment's maxtrialnum.     
+
+	     * Usage
+
+		       nnictl update trialnum [OPTIONS] 
+
+            Options:
+
+            | Name, shorthand | Required|Default | Description |
+            | ------ | ------ | ------ |------ |
+           | id|  False| |ID of the experiment you want to set|
+           | --value, -v|  True| |the new number of maxtrialnum you want to set|
+
      
 
 * __nnictl trial__

docs/SetupNNIDeveloperEnvironment.md

@@ -47,7 +47,7 @@ And open WebUI to check if everything is OK
 
 **6. Redeploy**
 
-After you change some code, just use **step 4** to rebuild your code, then the change will take effect immediately
+After the code changes, use **step 3** to rebuild your codes, then the changes will take effect immediately.
 
 ---
 At last, wish you have a wonderful day.

examples/trials/mnist-annotation/mnist.py

@@ -157,8 +157,8 @@ def main(params):
     '''
     # Import data
     mnist = input_data.read_data_sets(params['data_dir'], one_hot=True)
-    print('Mnist download data down.')
-    logger.debug('Mnist download data down.')
+    print('Mnist download data done.')
+    logger.debug('Mnist download data done.')
 
     # Create the model
     # Build the graph for the deep net
@@ -180,15 +180,15 @@ def main(params):
     test_acc = 0.0
     with tf.Session() as sess:
         sess.run(tf.global_variables_initializer())
-        """@nni.variable(nni.choice(50, 250, 500), name=batch_num)"""
-        batch_num = params['batch_num']
-        for i in range(batch_num):
-            batch = mnist.train.next_batch(batch_num)
-            """@nni.variable(nni.choice(1, 5), name=dropout_rate)"""
+        """@nni.variable(nni.choice(1, 4, 8, 16, 32), name=batch_size)"""
+        batch_size = params['batch_size']
+        for i in range(params['batch_num']):
+            batch = mnist.train.next_batch(batch_size)
+            """@nni.variable(nni.choice(0.5, 0.9), name=dropout_rate)"""
             dropout_rate = params['dropout_rate']
             mnist_network.train_step.run(feed_dict={mnist_network.images: batch[0],
                                                     mnist_network.labels: batch[1],
-                                                    mnist_network.keep_prob: dropout_rate}
+                                                    mnist_network.keep_prob: 1 - dropout_rate}
                                         )
 
             if i % 100 == 0:
@@ -224,7 +224,8 @@ def generate_defualt_params():
         'pool_size': 2,
         'hidden_size': 1024,
         'learning_rate': 1e-4,
-        'batch_num': 200}
+        'batch_num': 2000,
+        'batch_size': 32}
     return params
 
 

examples/trials/mnist-smartparam/mnist.py

@@ -180,13 +180,13 @@ def main(params):
     test_acc = 0.0
     with tf.Session() as sess:
         sess.run(tf.global_variables_initializer())
-        batch_num = nni.choice(50, 250, 500, name='batch_num')
-        for i in range(batch_num):
-            batch = mnist.train.next_batch(batch_num)
-            dropout_rate = nni.choice(1, 5, name='dropout_rate')
+        batch_size = nni.choice(1, 4, 8, 16, 32, name='batch_size')
+        for i in range(2000):
+            batch = mnist.train.next_batch(batch_size)
+            dropout_rate = nni.choice(0.5, 0.9, name='dropout_rate')
             mnist_network.train_step.run(feed_dict={mnist_network.images: batch[0],
                                                     mnist_network.labels: batch[1],
-                                                    mnist_network.keep_prob: dropout_rate}
+                                                    mnist_network.keep_prob: 1 - dropout_rate}
                                         )
 
             if i % 100 == 0:
@@ -223,6 +223,7 @@ def generate_defualt_params():
 
 if __name__ == '__main__':
     try:
+        nni.get_next_parameter()
         main(generate_defualt_params())
     except Exception as exception:
         logger.exception(exception)

examples/trials/mnist/config_assessor.yml

@@ -2,10 +2,10 @@ authorName: default
 experimentName: example_mnist
 trialConcurrency: 1
 maxExecDuration: 1h
-maxTrialNum: 1
+maxTrialNum: 20
 #choice: local, remote
 trainingServicePlatform: local
-searchSpacePath: ~/nni/examples/trials/mnist/search_space.json
+searchSpacePath: search_space.json
 #choice: true, false
 useAnnotation: false
 tuner:
@@ -23,5 +23,5 @@ assessor:
     optimize_mode: maximize
 trial:
   command: python3 mnist.py
-  codeDir: ~/nni/examples/trials/mnist
+  codeDir: .
   gpuNum: 0

examples/trials/mnist/mnist.py

@@ -173,10 +173,10 @@ def main(params):
     with tf.Session() as sess:
         sess.run(tf.global_variables_initializer())
         for i in range(params['batch_num']):
-            batch = mnist.train.next_batch(params['batch_num'])
+            batch = mnist.train.next_batch(params['batch_size'])
             mnist_network.train_step.run(feed_dict={mnist_network.images: batch[0],
                                                     mnist_network.labels: batch[1],
-                                                    mnist_network.keep_prob: params['dropout_rate']}
+                                                    mnist_network.keep_prob: 1 - params['dropout_rate']}
                                         )
 
             if i % 100 == 0:
@@ -212,7 +212,8 @@ def generate_default_params():
         'pool_size': 2,
         'hidden_size': 1024,
         'learning_rate': 1e-4,
-        'batch_num': 200}
+        'batch_num': 2000,
+        'batch_size': 32}
     return params
 
 

examples/trials/mnist/search_space.json

@@ -1,6 +1,7 @@
 {
-    "dropout_rate":{"_type":"uniform","_value":[0.1,0.5]},
+    "dropout_rate":{"_type":"uniform","_value":[0.5, 0.9]},
     "conv_size":{"_type":"choice","_value":[2,3,5,7]},
     "hidden_size":{"_type":"choice","_value":[124, 512, 1024]},
+    "batch_size": {"_type":"choice", "_value": [1, 4, 8, 16, 32]},
     "learning_rate":{"_type":"choice","_value":[0.0001, 0.001, 0.01, 0.1]}
 }

setup.py

@@ -35,7 +35,7 @@ def run(self):
 
 setup(
     name = 'nni',
-    version = '0.3.2',
+    version = '0.3.4',
     author = 'Microsoft NNI Team',
     author_email = '[email protected]',
     description = 'Neural Network Intelligence project',

src/nni_manager/rest_server/restValidationSchemas.ts

@@ -94,7 +94,9 @@ export namespace ValidationSchemas {
             params: joi.object(STARTEXPERIMENT.body).required(),
             execDuration: joi.number().required(),
             startTime: joi.number(),
-            endTime: joi.number()
+            endTime: joi.number(),
+            logDir: joi.string(),
+            maxSequenceId: joi.number()
         }
     };
     export const STARTTENSORBOARD = {

src/nni_manager/training_service/local/localTrainingService.ts

@@ -78,7 +78,7 @@ class LocalTrialJobDetail implements TrialJobDetail {
     public pid?: number;
 
     constructor(id: string, status: TrialJobStatus, submitTime: number,
-                workingDirectory: string, form: JobApplicationForm, sequenceId: number) {
+        workingDirectory: string, form: JobApplicationForm, sequenceId: number) {
         this.id = id;
         this.status = status;
         this.submitTime = submitTime;
@@ -291,21 +291,21 @@ class LocalTrainingService implements TrainingService {
     public getClusterMetadata(key: string): Promise<string> {
         switch (key) {
             case TrialConfigMetadataKey.TRIAL_CONFIG:
-                let getResult : Promise<string>;
-                if(!this.localTrailConfig) {
+                let getResult: Promise<string>;
+                if (!this.localTrailConfig) {
                     getResult = Promise.reject(new NNIError(NNIErrorNames.NOT_FOUND, `${key} is never set yet`));
                 } else {
-                    getResult = Promise.resolve(!this.localTrailConfig? '' : JSON.stringify(this.localTrailConfig));
+                    getResult = Promise.resolve(!this.localTrailConfig ? '' : JSON.stringify(this.localTrailConfig));
                 }
-                return getResult;     
+                return getResult;
             default:
                 return Promise.reject(new NNIError(NNIErrorNames.NOT_FOUND, 'Key not found'));
         }
     }
 
     public cleanUp(): Promise<void> {
         this.stopping = true;
-        for(const stream of this.streams) {
+        for (const stream of this.streams) {
             stream.destroy();
         }
         return Promise.resolve();
@@ -367,7 +367,7 @@ class LocalTrainingService implements TrainingService {
         await cpp.exec(`mkdir -p ${trialJobDetail.workingDirectory}`);
         await cpp.exec(`mkdir -p ${path.join(trialJobDetail.workingDirectory, '.nni')}`);
         await cpp.exec(`touch ${path.join(trialJobDetail.workingDirectory, '.nni', 'metrics')}`);
-        await fs.promises.writeFile(path.join(trialJobDetail.workingDirectory, 'run.sh'), runScriptLines.join('\n'), { encoding: 'utf8' });
+        await fs.promises.writeFile(path.join(trialJobDetail.workingDirectory, 'run.sh'), runScriptLines.join('\n'), { encoding: 'utf8', mode: 0o777 });
         await this.writeParameterFile(trialJobDetail.workingDirectory, (<TrialJobApplicationForm>trialJobDetail.form).hyperParameters);
         await this.writeSequenceIdFile(trialJobId);
         const process: cp.ChildProcess = cp.exec(`bash ${path.join(trialJobDetail.workingDirectory, 'run.sh')}`);

src/sdk/pynni/nni/trial.py

@@ -47,6 +47,7 @@ def get_next_parameter():
     return _params['parameters']
 
 def get_current_parameter(tag):
+    global _params
     if _params is None:
         return None
     return _params['parameters'][tag]