brancher

Domain: forums.fast.ai

RuntimeError: invalid argument 0: Sizes of tensors must match except in dimension 0

RuntimeError: invalid argument 0: Sizes of tensors must match except in dimension 0

Environment information fastai version: 1.0.38 PyTorch version: 1.0.0 Is debug build: No CUDA used to build PyTorch: 9.0.176 OS: Ubuntu 16.04.5 LTS GCC version: (Ubuntu 5.4.0-6ubuntu1~16.04.11) 5.4.0 20160609 CMake version: version 3.12.2 Python version: 3.6 Is CUDA available: Yes CUDA runtime version: Could not collect GPU models and configuration: GPU 0: Quadro M4000 Nvidia driver version: 410.48 cuDNN version: Could not collect Versions of relevant libraries: [pip] Could not col...

favicon
fast.ai
Unofficial pytorch 0.4 support

Unofficial pytorch 0.4 support

Thanks to @sgugger for fixing the last issues in AWD LSTM, you should now find that all of parts 1 and 2 of the course run fine under pytorch 0.4 (the new version that was just released). There’s no need to upgrade however, and we’re not updating environment.yml to push the new version - but if you have to upgrade for some other reason, things should work fine. (If they don’t, feel free to ask for help here, but I can’t promise to make fixing it a priority, since I’m not considering 0.4 “offici...

favicon
fast.ai
Reset GPU without restarting linux?

Reset GPU without restarting linux?

I setup this computer to use remotely. In some instances CUDA errors (maybe related to network issues, I can’t tell) left the GPU useless. Killing the jupyter kernel didn’t help, only a computer restart. This is the only GPU in the system (1070ti), so I believe it’s in use by the display. I am not running xwindows or similar. nvidia-smi reset doesn’t seem to help either: tbatchelli@MLrig:~$ nvidia-smi -r GPU Reset couldn't run because GPU 00000000:23:00.0 is the primary GPU. Is there any way...

favicon
fast.ai
Jupyter Notebook - how to enable Intellisense

Jupyter Notebook - how to enable Intellisense

Hello, I’m new to Python and when using the notebooks I missed intellisense (enables you to ask an object what methods it has). At the top of your note book add this line %config IPCompleter.greedy=True Then when you have an object, for example numpy (np) do this np. After the . press [TAB] and it will show you all the methods available. Another useful thing is method parameters. Juypter will show you these if you press [SHIFT] and [TAB] from within the method parentheses. Eg. Cheers ...

favicon
fast.ai
(gcloud.compute.ssh) Could not SSH into the instance

(gcloud.compute.ssh) Could not SSH into the instance

hello everyone ! before posting, i tried to find other posts whith the same problem i face but i did not find, so i hope it will not be redundant with an other post ! After having saw first video on image classification, i started to set up my Google Cloud Platform (GCP) environment. I release each step succesfully expect until command : gcloud compute ssh --zone=$ZONE jupyter@$INSTANCE_NAME -- -L 8080:localhost:8080 Each times, i have the same error results. I used ubuntu App for Windows 1...

favicon
fast.ai
Help wanted (easy one!) - clean up imagenet validation set

Help wanted (easy one!) - clean up imagenet validation set

Turns out there’s a blacklist of imagenet validation files here: https://github.com/dailuo/create_dataset_imagenet/blob/master/ILSVRC2014_clsloc_validation_blacklist.txt There should be a file called ILSVRC2012_val_{i}.JPEG in one of the validation category folders, where {i} is zero pre-padded, e.g. ILSVRC2012_val_00045880.JPEG. I wonder if someone could be so kind as to create a list of path and file names for all these images, so we can easily remove them from our training? You’ll need a co....

favicon
fast.ai
Question of tensorflow : How could I turn is_training of batchnorm to False

Question of tensorflow : How could I turn is_training of batchnorm to False

net = tf.layers.conv2d(inputs = features, filters = 64, kernel_size = [3, 3], strides = (2, 2), padding = 'same') net = tf.contrib.layers.batch_norm(net, is_training = True) net = tf.nn.relu(net) net = tf.reshape(net, [-1, 64 * 7 * 7]) # net = tf.layers.dense(inputs = net, units = class_num, kernel_initializer = tf.contrib.layers.xavier_initializer(), name = 'regression_output') #...... #after training saver = tf.train.Saver() saver.save(sess, 'reshape_final.ckpt') tf.train.write_graph(sess.gr....

favicon
fast.ai
Successful ubuntu 18.04 with iGPU for xserver and nvidia GPU for CUDA work setup

Successful ubuntu 18.04 with iGPU for xserver and nvidia GPU for CUDA work setup

FYI, I have just successfully installed a fresh Kubuntu 18.04.1 with CUDA 9.2 in a dual setup - onboard intel GPU for the screen, and Nvidia GPU for ML CUDA work. All using apt packages. You can see all the details here, including a resolution for a hiccup with trying to get xorg to run on iGPU, instead of the Nvidia GPU. https://askubuntu.com/questions/1061551/configuring-igpu-for-xserver-and-nvidia-gpu-for-cuda-work-kubuntu-18-04-cuda-9 Both fastai and fastai_v1 run on it w/o any problems. I....

favicon
fast.ai
VGG16 Change Size from (224,224) or Change Images size (256,256)

VGG16 Change Size from (224,224) or Change Images size (256,256)

Is it better to use the default size of (224,224) and change my images to that size (I’m trying to modify the dogs and cats to compete here: https://www.hackerearth.com/challenge/competitive/deep-learning-challenge-1/) or modify vgg16 to use images of size 256x256. What kind of consequences does that have on the model? I have tried using the vgg16 model from Keras and I am not getting great results, but I believe part of the reason is that they aren’t letting me use the imagenet images to give....

favicon
fast.ai
Very volatile validation loss

Very volatile validation loss

Hi, I’m training a dense CNN model and noticed that If I pick too high of a learning rate I get better validation results (as picked up by model checkpoint) than If I pick a lower learning rate. The problem is that high learning rate gives wild fluctuations in validation loss, while training with low learning rates provides for a smooth and pleasing to an eye learning curve which never goes as low though. My train set is around 900 images and test set is close to 100. I do augmentation on the ....

favicon
fast.ai
Notadirectoryerror: [winerror 267] the directory name is invalid:
Minimap

Notadirectoryerror: [winerror 267] the directory name is invalid:

I'm running the Bitnami Django Stack (installed on F: Hard disk Drive and not on C : ) and I want to collect static files and compress them into a single location by running python manage.py collectstatic --traceback, but when I run the command I get this error: WindowsError: [Error 267] The directory name is invalid I have read on Microsoft Support that I need to change the properties of the Command Prompt (cmd) by changing %HOMEDRIVE%%HOMEPATH% to %WINDIR% and I did the same for Bitname Con..

0
7