log

docs/source/index.rst

@@ -13,7 +13,64 @@ Welcome to jdit's documentation!
    model
    optimizer
    trainer
-   metric
+
+
+
+**Jdit** is a research processing oriented framework based on pytorch. Only care about your ideas.
+You don't need to build a long boring code to run a deep learning project to verify your ideas.
+
+You only need to implement you ideas and
+don't do anything with training framework, multiply-gpus, checkpoint, process visualization, performance evaluation and so on.
+
+Quick start
+-----------
+After building and installing jdit package, you can make a new directory for a quick test.
+Assuming that you get a new directory `example`.
+run this code in `ipython` cmd.(Create a `main.py` file is also acceptable.)
+
+.. code:: python
+
+    from jdit.trainer.instances.fashingClassification
+    import start_example
+    start_example()
+
+Then you will see something like this as following.
+
+.. code:: python
+
+    ===> Build dataset
+    use 8 thread!
+    Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz
+    Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz
+    Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz
+    Downloading http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz
+    Processing...
+    Done!
+    ===> Building model
+    ResNet Total number of parameters: 2776522
+    ResNet model use CPU!
+    apply kaiming weight init!
+    ===> Building optimizer
+    ===> Training
+    using `tensorboard --logdir=log` to see learning curves and net structure.
+    training and valid data, configures info and checkpoint were save in `log` directory.
+      0%|            | 0/10 [00:00<?, ?epoch/s]
+    0step [00:00, ?step/s]
+
+* It will search a fashing mnist dataset.
+* Then build a resnet18 for classification.
+* For training process, you can find learning curves in `tensorboard`.
+* It will create a `log` directory in `example/`, which saves training processing data and configures.
+
+Although it is just an example, you still can build your own project easily by using jdit framework.
+Jdit framework can deal with
+* Data visualization. (learning curves, images in pilot process)
+* CPU, GPU or GPUs. (Training your model on specify devices)
+* Intermediate data storage. (Saving training data into a csv file)
+* Model checkpoint automatically.
+* Flexible templates can be used to integrate and custom overrides.
+So, let's see what is **jdit**.
+
 
 Indices and tables
 ==================

docs/source/trainer.rst

@@ -13,6 +13,12 @@ SupTrainer
 .. autoclass:: SupTrainer
     :members:
 
+ClassificationTrainer
+-------------------
+
+.. autoclass:: ClassificationTrainer
+    :members:
+
 GanTrainer
 ----------
 

jdit/trainer/classification.py

@@ -1,14 +1,13 @@
-# import torch
-# from torch.autograd import Variable
 from torch.nn import CrossEntropyLoss
 from .super import *
 from abc import abstractmethod
 from tqdm import *
 
-
 class ClassificationTrainer(SupTrainer):
-    num_class = None
+    """this is a classification trainer.
 
+    """
+    num_class = None
     def __init__(self, logdir, nepochs, gpu_ids, net, opt, datasets):
         super(ClassificationTrainer, self).__init__(nepochs, logdir, gpu_ids_abs=gpu_ids)
         self.net = net

jdit/trainer/instances/fashingClassification.py

@@ -10,6 +10,9 @@
 
 
 class FashingClassTrainer(ClassificationTrainer):
+    """this is an instance of how to use `ClassificationTrainer` to build your own trainer
+
+    """
     mode = "L"
     num_class = 10
     every_epoch_checkpoint = 20  # 2
@@ -55,6 +58,10 @@ def compute_valid(self):
 
 
 def start_example():
+    """ run this to test a `FashingClassTrainer` instance
+
+    :return:
+    """
     gpus = [0]
     batchSize = 64
     nepochs = 10

jdit/trainer/super.py

@@ -12,6 +12,9 @@
 
 
 class SupTrainer(object):
+    """this is a super class of all trainers
+
+    """
     every_epoch_checkpoint = 10
     every_epoch_changelr = 0
     mode = "L"
@@ -112,6 +115,9 @@ def configure(self):
 
 
 class Loger(object):
+    """this is a log recorder.
+
+    """
     def __init__(self, logdir="log"):
         self.logdir = logdir
         self.regist_list = []
@@ -183,6 +189,9 @@ def clear_regist(self):
 
 
 class Watcher(object):
+    """this is a params and images watcher
+
+    """
     def __init__(self, logdir, mode="L"):
         self.logdir = logdir
         self.writer = SummaryWriter(log_dir=logdir)
@@ -280,6 +289,9 @@ def _buildDir(self, dirs):
 
 
 class Performance(object):
+    """this is a performance watcher.
+
+    """
 
     def __init__(self, gpu_ids_abs=()):
         self.config_dic = dict()