refine README for 0.1.4 version.

docs/keras2cntk.md

@@ -52,7 +52,7 @@ $ mmconvert -sf keras -iw imagenet_inception_v3.h5 -df cntk -om keras_to_cntk_in
 CNTK model file is saved as [keras_to_cntk_inception_v3.dnn], generated by [2c33f7f278cb46be992f50226fcfdb5d.py] and [2c33f7f278cb46be992f50226fcfdb5d.npy].
 ```
 
-Then you get the CNTK original model *keras_to_cntk_inception_v3.dnn* converted from Keras.
+Then you get the CNTK original model *keras_to_cntk_inception_v3.dnn* converted from Keras. **2c33f7f278cb46be992f50226fcfdb5d.py** and **2c33f7f278cb46be992f50226fcfdb5d.npy** are temporal files which will be removed automatically.
 
 ### **Step-by-step Command (for debugging)**
 

docs/tf2pytorch.md

@@ -6,12 +6,14 @@ Source: TensorFlow
 
 Destination: PyTorch
 
-0. Prepare the TensorFlow model
+---
+
+## Prepare the TensorFlow model
 
 You need to prepare your pre-trained TensorFlow model firstly. And there is a pre-trained model extractor for frameworks to help you. You can refer it to extract your TensorFlow model checkpoint files.
 
 ```bash
-$ python -m mmdnn.conversion._script.extractModel -f tensorflow -n resnet_v2_152
+$ mmdownload -f tensorflow -n resnet_v2_152
 
 Downloading file [./resnet_v2_152_2017_04_14.tar.gz] from [http://download.tensorflow.org/models/resnet_v2_152_2017_04_14.tar.gz]
 100% [......................................................................] 675629399 / 675629399
@@ -21,12 +23,12 @@ Model saved in file: ./imagenet_resnet_v2_152.ckpt
 
 The you got the TensorFlow checkpoint files for *ResNet V2 152* model which is in your current working directory, including *imagenet_resnet_v2_152.ckpt.meta* for architecture , *imagenet_resnet_v2_152.ckpt.data-00000-of-00001*  and *imagenet_resnet_v2_152.ckpt.index* for weights.
 
-1. Find the output node of the model
+## Find the output node of the model
 
 TensorFlow original checkpoint files contain many operators (if you tried tensorboard to visualize the graph) which is not used in our toolkits. we should prune them with specifying the output node of your model.
 
 ```bash
-$ python -m mmdnn.conversion.examples.tensorflow.vis_meta imagenet_resnet_v2_152.ckpt.meta ./log/
+$ mmvismeta imagenet_resnet_v2_152.ckpt.meta ./log/
 .
 .
 .
@@ -39,10 +41,31 @@ The you can open URL above to find the output node of your model,
 
 like the squeeze node named *MMdnn_Output* we setup in our tensorflow model extractor. Detail information is in [TensorFlow README](https://github.com/Microsoft/MMdnn/blob/master/mmdnn/conversion/tensorflow/README.md)
 
-2. Convert the pre-trained model files to intermediate representation
+## Convert TensorFlow Model to PyTorch
+
+We provide two ways to convert models.
+
+### **One-step Command**
+
+Above MMdnn@0.1.4, we provide one command to achieve the conversion
+
+```bash
+$ mmconvert -sf tensorflow -in imagenet_resnet_v2_152.ckpt.meta -iw imagenet_resnet_v2_152.ckpt --dstNode MMdnn_Output -df pytorch -om tf_to_pytorch_resnet_152.pth
+.
+.
+.
+PyTorch model file is saved as [tf_to_pytorch_resnet_152.pth], generated by [052eb72db9934edc90d8e1ffa48144d7.py] and [052eb72db9934edc90d8e1ffa48144d7.npy].
+```
+
+Then you get the PyTorch original model *tf_to_pytorch_resnet_152.pth* converted from TensorFlow. **052eb72db9934edc90d8e1ffa48144d7.py** and **052eb72db9934edc90d8e1ffa48144d7.npy** are temporal files which will be removed automatically.
+
+
+### Step-by-step Command for debugging
+
+#### Convert the pre-trained model files to intermediate representation
 
 ```bash
-$ python -m mmdnn.conversion._script.convertToIR -f tensorflow -n imagenet_resnet_v2_152.ckpt.meta -w imagenet_resnet_v2_152.ckpt --dstNode MMdnn_Output -o converted
+$ mmtoir -f tensorflow -n imagenet_resnet_v2_152.ckpt.meta -w imagenet_resnet_v2_152.ckpt --dstNode MMdnn_Output -o converted
 
 Parse file [imagenet_resnet_v2_152.ckpt.meta] with binary format successfully.
 Tensorflow model file [imagenet_resnet_v2_152.ckpt.meta] loaded successfully.
@@ -54,10 +77,10 @@ IR weights are saved as [converted.npy].
 
 Then you got the **intermediate representation** files *converted.json* for visualization, *converted.proto* and *converted.npy* for next steps.
 
-3. Convert the IR files to PyTorch code
+#### Convert the IR files to PyTorch code
 
 ```bash
-$ python -m mmdnn.conversion._script.IRToCode -f pytorch -n converted.pb -w converted.npy -d converted_pytorch.py -dw converted_pytorch.npy
+$ mmtocode -f pytorch -n converted.pb -w converted.npy -d converted_pytorch.py -dw converted_pytorch.npy
 
 Parse file [converted.pb] with binary format successfully.
 Target network code snippet is saved as [converted_pytorch.py].
@@ -68,7 +91,7 @@ And you will get a filename *converted_pytorch.py*, which contains the **origina
 
 With the three steps, you have already converted the pre-trained TensorFlow *ResNet V2 152* models to PyTorch network building file *converted_pytorch.py* and weights file *converted_pytorch.npy*. You can use these two files to fine-tune training or inference.
 
-4. Dump the original PyTorch model
+#### Dump the original PyTorch model
 
 ```bash
 $ python -m mmdnn.conversion.examples.pytorch.imagenet_test -n converted_pytorch.py -w converted_pytorch.npy --dump converted_pytorch.pth

mmdnn/conversion/examples/tensorflow/extractor.py

@@ -155,11 +155,11 @@ def handle_checkpoint(cls, architecture, path):
 
         init = tf.global_variables_initializer()
         with tf.Session() as sess:
-            writer = tf.summary.FileWriter('./graphs', sess.graph)
+            # tf.train.export_meta_graph("kit.meta", as_text=True)
+            # writer = tf.summary.FileWriter('./graphs', sess.graph)
             writer.close()
             sess.run(init)
             saver = tf.train.Saver()
-            tf.train.export_meta_graph("kit.meta", as_text=True)
             saver.restore(sess, path + cls.architecture_map[architecture]['filename'])
             save_path = saver.save(sess, path + "imagenet_{}.ckpt".format(architecture))
             print("Model saved in file: %s" % save_path)

mmdnn/conversion/tensorflow/README.md

@@ -19,7 +19,7 @@ Support frameworks: ['resnet_v1_152', 'inception_v3', 'resnet_v2_50', 'resnet_v2
 Example:
 
 ```bash
-$ python -m mmdnn.conversion._script.extractModel -f tensorflow -n resnet_v2_152
+$ mmdownload -f tensorflow -n resnet_v2_152
 
 Downloading file [./resnet_v2_152_2017_04_14.tar.gz] from [http://download.tensorflow.org/models/resnet_v2_152_2017_04_14.tar.gz]
 100% [......................................................................] 675629399 / 675629399
@@ -30,7 +30,6 @@ Model saved in file: ./imagenet_resnet_v2_152.ckpt
 
 Then you can see files *imagenet_resnet_v2_152.ckpt.data-00000-of-00001*, *imagenet_resnet_v2_152.ckpt.index* and *imagenet_resnet_v2_152.ckpt.meta*, which can be handled by Tensorflow parser.
 
-
 Mainly extract code like:
 
 ```python
@@ -48,7 +47,7 @@ with slim.arg_scope(...):
 When you prepared your checkpoint, you can find the output node name from the graph by Tensorboard.
 
 ```bash
-$ python -m mmdnn.conversion.examples.tensorflow.vis_meta imagenet_resnet_v2_152.ckpt.meta ./logs/
+$ mmvismeta imagenet_resnet_v2_152.ckpt.meta ./logs/
 
 TensorBoard 0.4.0rc3 at http://kit-station:6006 (Press CTRL+C to quit)
 ```