I am a passionate software developer interested/experienced in building (micro)services and a background in deep learning and computer vision.
Portfolio: https://abhishekrana.github.io
My GitHub Stats
Top Repositories
Apparel detection using deep learning
License: Apache License 2.0
I am a passionate software developer interested/experienced in building (micro)services and a background in deep learning and computer vision.
Portfolio: https://abhishekrana.github.io
My GitHub Stats
Top Repositories
Hey,
I am trying to train a fashion Object Detector. According to the steps given in your readme file, I first have to use dataset_create.py
Now as per my understanding that creates many cropped images out of each image via selective search. So, after running that script, in my dataset folder, in train, test and val, there are many cropped images. Many of these images have just the head of the person wearing those clothes and many other wrong images. So, won't such images confuse our object detector when we train?
I need your help to understand how this selective search works and how it is useful?
Thanks
--> python train.py
Using TensorFlow backend.
tf.estimator package not installed.
tf.estimator package not installed.
[ 23 : init ] batch_size 32
[ 77 : get_subdir_list ] names ['Coat', 'Kaftan', 'Robe']
[ 27 : init ] class_names ['Coat', 'Kaftan', 'Robe']
[ 31 : init ] input_shape (224, 224, 3)
[ 51 : save_bottleneck ] class_names ['Coat', 'Kaftan', 'Robe']
[ 52 : save_bottleneck ] batch_size 32
[ 53 : save_bottleneck ] epochs 100
[ 54 : save_bottleneck ] input_shape (224, 224, 3)
2019-01-22 07:38:22.153964: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:964] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-01-22 07:38:22.154485: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1432] Found device 0 with properties:
name: Tesla K80 major: 3 minor: 7 memoryClockRate(GHz): 0.8235
pciBusID: 0000:00:04.0
totalMemory: 11.17GiB freeMemory: 11.10GiB
2019-01-22 07:38:22.154524: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1511] Adding visible gpu devices: 0
2019-01-22 07:38:22.496875: I tensorflow/core/common_runtime/gpu/gpu_device.cc:982] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-01-22 07:38:22.496937: I tensorflow/core/common_runtime/gpu/gpu_device.cc:988] 0
2019-01-22 07:38:22.496957: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1001] 0: N
2019-01-22 07:38:22.497239: W tensorflow/core/common_runtime/gpu/gpu_bfc_allocator.cc:42] Overriding allow_growth setting because the TF_FORCE_GPU_ALLOW_GROWTH environment variable is set. Original config value was 0.
2019-01-22 07:38:22.497324: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 10758 MB memory) -> physical GPU (device: 0, name: Tesla K80, pci bus id: 0000:00:04.0, compute capability: 3.7)
[ 67 : save_bottleneck ] dataset_train_class_path dataset/train/Coat
[ 73 : save_bottleneck ] images_path_name 0
[ 67 : save_bottleneck ] dataset_train_class_path dataset/train/Kaftan
[ 73 : save_bottleneck ] images_path_name 0
[ 67 : save_bottleneck ] dataset_train_class_path dataset/train/Robe
[ 73 : save_bottleneck ] images_path_name 0
[ 67 : save_bottleneck ] dataset_train_class_path dataset/validation/Coat
[ 73 : save_bottleneck ] images_path_name 0
[ 67 : save_bottleneck ] dataset_train_class_path dataset/validation/Kaftan
[ 73 : save_bottleneck ] images_path_name 0
[ 67 : save_bottleneck ] dataset_train_class_path dataset/validation/Robe
[ 73 : save_bottleneck ] images_path_name 0
[ 154 : train_model ] train_labels_iou []
[ 155 : train_model ] train_labels_iou <class 'numpy.ndarray'>
[ 156 : train_model ] train_labels_class <class 'numpy.ndarray'>
[ 157 : train_model ] train_labels_class (0,)
Traceback (most recent call last):
File "train.py", line 277, in
train_model()
File "train.py", line 166, in train_model
n1, n2, w, h, c = train_data.shape
ValueError: not enough values to unpack (expected 5, got 1)
I need evaluate model with mAP but I can't write it. Please help me
The dataset is downloaded and organized as per the request. Also, dataset_create returns no error, but an error is encountered while train.py which is not enough value to unpack. Please help
Hi,
Recently I download your script and it works (Training set includes ['coat', 'kaftan', 'robe']).
But when I add ['Chino'] in the training set, and run dataset_create.py, it will meet a problem.
It will stop at a certain image and display an error.
I will meet the same problem if I add ['Jeans'] to the training set.
It seems some certain images will cause this issue.
I don't know if I should do some changes to the ‘crop function’. I think I need your help.
Thanks a lot!
I don't know why it's giving me dimension error, during prediction?
Traceback (most recent call last):
File "predict.py", line 225, in
predict_model(image_crops, image_crops_name)
File "predict.py", line 54, in predict_model
model = create_model(False, True, input_shape, len(class_names), optimizer, learn_rate, decay, momentum, activation, dropout_rate)
File "/home/prashant/DeepFashion/model.py", line 67, in create_model
model.load_weights(top_model_weights_path_load, by_name=True)
File "/usr/local/lib/python2.7/dist-packages/keras/engine/topology.py", line 2570, in load_weights
load_weights_from_hdf5_group_by_name(f, self.layers)
File "/usr/local/lib/python2.7/dist-packages/keras/engine/topology.py", line 3111, in load_weights_from_hdf5_group_by_name
K.batch_set_value(weight_value_tuples)
File "/usr/local/lib/python2.7/dist-packages/keras/backend/tensorflow_backend.py", line 2183, in batch_set_value
assign_op = x.assign(assign_placeholder)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/variables.py", line 594, in assign
return state_ops.assign(self._variable, value, use_locking=use_locking)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/state_ops.py", line 276, in assign
validate_shape=validate_shape)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/gen_state_ops.py", line 59, in assign
use_locking=use_locking, name=name)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/op_def_library.py", line 787, in _apply_op_helper
op_def=op_def)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 3162, in create_op
compute_device=compute_device)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 3208, in _create_op_helper
set_shapes_for_outputs(op)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 2427, in set_shapes_for_outputs
return _set_shapes_for_outputs(op)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 2400, in _set_shapes_for_outputs
shapes = shape_func(op)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 2330, in call_with_requiring
return call_cpp_shape_fn(op, require_shape_fn=True)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/common_shapes.py", line 627, in call_cpp_shape_fn
require_shape_fn)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/common_shapes.py", line 691, in _call_cpp_shape_fn_impl
raise ValueError(err.message)
ValueError: Dimension 1 in both shapes must be equal, but are 19 and 3. Shapes are [256,19] and [256,3]. for 'Assign_32' (op: 'Assign') with input shapes: [256,19], [256,3].
please publish a license, preferably apache 2.0
What was the accuracy that you were able to get for the class predictions?
thank you very much for your project,it's very help.but i have problem about the classification of attributes to ask for advice,class such as color,style,pattern;how can i do to finish this trough this project? Your help is really appreciated!
Has anyone fixed the TODOs in the train_model function regarding the class weight value calculation and the best weights?
When I try to install requirements from txt it cannot find opencv==1.0.1 I looked up pypi doesnt shot package named as opencv, is it from another source?
Sir,
I read your repo and went through the code, found it really interesting and looking forward to use it in my current project. But when I downloaded the dataset ,I don't have enough specs on my system to meet the requirements. I request you to please share the trained model file since I can't afford that king of system at this age.
Please help me with this.
Thank you a lot sir,
Cheers
训练时间过长,已经训练好几天了,还没有搞完
Is there any pre-trained model readily available ?
Do you have a link where we can download the pre-trained Keras models on DeepFashion - like the ones ones referred to in the config.py file ?
Hi Sir,
According to the installation of the python package in your , some packages cannot be installed, especially the opencv1.0.1. There is no such version on the official website,how can i solve the problem. Does the versions of the python packages must be installed as you list in requirements.txt?
I have an issue when running train.py, which is the job will be killed while running this part of creating train set: (I have chosen ten classes of data, which is not so much.)
train_data = []
for index, btl_name in enumerate(btl_train_names):
temp = np.load(open(btl_name))
train_data.append(temp)
I am using a server with 256GB memory and one NVIDIA GPU. I want to know the minimum configuration for running the code successfully. Anyone could help me ??
thanks,
Seems that the data set is already deleted on Dropbox. It there any other way to get it?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.