mberkay0 / clothing-segmentation

Clothing segmentation with DeepLabV3+

Hi there, I'm Berkay 👋

About me 🔭

• 🖥 I'm a full-stack ML & DL developer
• 🎓 I'm an Electrical and Electronics Engineer
• 💻 I love data science and deep learning
• 🌱 I’m currently learning everything about AI 🤔
• 👯 I’m looking to collaborate with other content creators
• ⚡ Fun fact: I love to programming
• 🌐 You can access all my open-source projects in this profile

Connect with me:

Languages and Tools:

---

🔔 My Recent GitHub Activity

---

⚡ Github Stats

 

💪 Maybe you want to star my projects

clothing-segmentation's Issues

Getting this issue while trying to load the weights. Please help

FileNotFoundError Traceback (most recent call last)
in <cell line: 3>()
weight_name = os.path.basename(w.name).replace(':0', '')
weight_file = layer.name + '_' + weight_name + '.npy'
---> weight_arr = np.load(os.path.join(WEIGHTS_DIR, weight_file))
weights.append(weight_arr)
layer.set_weights(weights)

/usr/local/lib/python3.10/dist-packages/numpy/lib/npyio.py in load(file, mmap_mode, allow_pickle, fix_imports, encoding, max_header_size)
own_fid = False
else:
--> fid = stack.enter_context(open(os_fspath(file), "rb"))
own_fid = True

FileNotFoundError: [Errno 2] No such file or directory: 'weights/mobilenetv2/Conv_kernel.npy'

InvalidArumentError error at startup of train

history = model.fit(train_gen,
                    validation_data=val_gen,
                    steps_per_epoch=len(train_gen),
                    validation_steps=len(val_gen),
                    epochs=num_epoch,
                    callbacks=callbacks,
                    verbose = 1)

2023-07-13 08:40:44.781595: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'Placeholder/_0' with dtype int32
[[{{node Placeholder/_0}}]]
Epoch 1/35
2023-07-13 08:40:45.341263: W tensorflow/core/framework/op_kernel.cc:1830] OP_REQUIRES failed at spacetobatch_op.cc:230 : INVALID_ARGUMENT: padded_shape[0]=32 is not divisible by block_shape[0]=6
2023-07-13 08:40:45.341315: W tensorflow/core/framework/op_kernel.cc:1830] OP_REQUIRES failed at spacetobatch_op.cc:230 : INVALID_ARGUMENT: padded_shape[0]=56 is not divisible by block_shape[0]=18
2023-07-13 08:40:45.341339: I tensorflow/core/common_runtime/executor.cc:1197] [/job:localhost/replica:0/task:0/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: padded_shape[0]=32 is not divisible by block_shape[0]=6
[[{{node deeplabv3plus/aspp1_depthwise/depthwise/SpaceToBatchND}}]]
2023-07-13 08:40:45.341346: W tensorflow/core/framework/op_kernel.cc:1830] OP_REQUIRES failed at spacetobatch_op.cc:230 : INVALID_ARGUMENT: padded_shape[0]=44 is not divisible by block_shape[0]=12

File "/root/anaconda3/envs/deeplabv3/lib/python3.11/site-packages/tensorflow/python/keras/backend.py", line 5574, in depthwise_conv2d
x = nn.depthwise_conv2d(
Node: 'deeplabv3plus/aspp1_depthwise/depthwise/SpaceToBatchND'
padded_shape[0]=32 is not divisible by block_shape[0]=6
[[{{node deeplabv3plus/aspp1_depthwise/depthwise/SpaceToBatchND}}]] [Op:__inference_train_function_64467]

Hi. Thanks for the great dataset.
I am getting the following error during the train.
How can I solve it??

NameError: name 'callbacks' is not defined

Pretty noob to ML, trying to get this project to run and getting this error. Can you help me out?

---

NameError Traceback (most recent call last)
in <cell line: 1>()
4 validation_steps=len(val_gen),
5 epochs=num_epoch,
----> 6 callbacks=callbacks,
7 verbose = 1)

NameError: name 'callbacks' is not defined

Pre-trained weights availability

Good afternoon and thank you very much for your work @mberkay0 😄

I am working in a personal project where your work would help me a lot, so are the pre-trained weights of the model available? If so, could you please share them with me? This would be very helpful, as I wouldn't need to train the model, given that you already did.

Best regards,
Dani