Problem in code

[Aristot1e]

When I ran the code 'train.py' encountered an unsolvable problem, so I came to ask for help. This error looks like this:

/home/img/anaconda3/envs/bin/python3.6 /home/img/Desktop/lff/MTrans-main/train.py doing sr_single DATASET: CHALLENGE: singlecoil MODE: train ROOT: /home/img/Desktop/lff/Dataset/single INPUT_SIZE: 160 MODEL: HEAD_HIDDEN_DIM: 16 INPUT_DIM: 1 OUTPUT_DIM: 1 PATCH_SIZE: 16 TRANSFORMER_DEPTH: 4 TRANSFORMER_MLP_RATIO: 3 TRANSFORMER_NUM_HEADS: 4 MULTI:

OUTPUTDIR: /home/img/Desktop/lff/MTrans-main/output/weights_SR_single_50x RESUME: SCALE: 2 SEED: 42 SOLVER: BATCH_SIZE: 8 DEVICE: cuda DEVICE_IDS: [0, 1] LR: 0.001 LR_DROP: [40] NUM_WORKERS: 16 PRINT_FREQ: 10 WEIGHT_DECAY: 0.0001 TEST_OUTPUTDIR: outputs/sr_single TRAIN: EPOCHS: 50 TRANSFORMS: ACCELERATIONS: [4] CENTER_FRACTIONS: [0.08] MASKTYPE: random USE_CL1_LOSS: False USE_MULTI_MODEL: False WORK_TYPE: sr dist_url: env:// world_size: 1 Not using distributed mode number of params: 641.73 M Traceback (most recent call last): File "/home/img/Desktop/lff/MTrans-main/train.py", line 170, in main(cfg, args.experiment) File "/home/img/Desktop/lff/MTrans-main/train.py", line 57, in main dataset_train = build_dataset(args, mode='train') File "/home/img/Desktop/lff/MTrans-main/data/fastmri.py", line 224, in build_dataset sample_rate=sample_rate, mode=mode) File "/home/img/Desktop/lff/MTrans-main/data/fastmri.py", line 119, in init with open(self.csv_file, 'r') as f: FileNotFoundError: [Errno 2] No such file or directory: '/home/img/Desktop/lff/Dataset/single/singlecoil_train/singlecoil_train_split_less.csv'

Process finished with exit code 1

I don’t know how to deal with it, so, please. Thanks for the help.

[Y445n9]

same problem, we have no csv file ,I wonder if author can provide this file?

[chunmeifeng]

Hi friends, you need to select the paired data by yourself.

Y445n9 @.***> 于2021年10月29日周五 上午11:33写道：

same problem, we have no csv file ,I wonder if author can provide this file?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-954595881, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN3TA4P2RIHGYEDUCK3UJJS6RANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[Aristot1e]

same problem, we have no csv file ,I wonder if author can provide this file?

Hi friend, have you solved this problem or reproduced this code? We can talk about this.

[chunmeifeng]

Hi friends, we have update the pytorch_lightning version as: pytorch_lightning==0.8.1

Aristot1e @.***> 于2021年11月16日周二 上午9:16写道：

same problem, we have no csv file ,I wonder if author can provide this file?

Hi friend, have you solved this problem or reproduced this code? We can talk about this.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-970000177, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN6JO4EKUGR5UG5CY4LUMIHMBANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[123wanglijuan]

Hi friends,how can select the paired data?What are the characteristics of the paired data

[chunmeifeng]

Hi, you can read the keys in the dictionary, and select the paired data according the keys

123wanglijuan @.***> 于2021年12月3日周五 08:18写道：

Hi friends,how can select the paired data?What are the characteristics of the paired data

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-985273392, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN2G2MO27T7S7XGMT7DUPBVKPANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[123wanglijuan]

Hi, you can read the keys in the dictionary, and select the paired data according the keys 123wanglijuan @.***> 于2021年12月3日周五 08:18写道： … Hi friends,how can select the paired data?What are the characteristics of the paired data — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN2G2MO27T7S7XGMT7DUPBVKPANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

Thank you very much,I read attributes of the HDF5 files, which can select the paired data according the keys

[chunmeifeng]

I don't remember the details, but I can give you an example: file1000685 and file1000568

123wanglijuan @.***> 于2021年12月5日周日 10:48写道：

Hi, you can read the keys in the dictionary, and select the paired data according the keys 123wanglijuan @.***> 于2021年12月3日周五 08:18写道： … <#m_3085210963525479724_> Hi friends,how can select the paired data?What are the characteristics of the paired data — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment) https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-985273392>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN2G2MO27T7S7XGMT7DUPBVKPANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub .

Thank you very much,I read attributes of the HDF5 files, which can select the paired data according the keys

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-986198192, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN6CSMAWPRGJVOXQJNDUPMYM7ANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[123wanglijuan]

I don't remember the details, but I can give you an example: file1000685 and file1000568 123wanglijuan @.> 于2021年12月5日周日 10:48写道： … Hi, you can read the keys in the dictionary, and select the paired data according the keys 123wanglijuan @.> 于2021年12月3日周五 08:18写道： … <#m_3085210963525479724_> Hi friends,how can select the paired data?What are the characteristics of the paired data — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment) <#1 (comment)>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN2G2MO27T7S7XGMT7DUPBVKPANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub . Thank you very much,I read attributes of the HDF5 files, which can select the paired data according the keys — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN6CSMAWPRGJVOXQJNDUPMYM7ANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

Thank you very much, I have selected the paired data based on your method.

[chunmeifeng]

Congratulations!

123wanglijuan @.***> 于2021年12月5日周日 13:31写道：

I don't remember the details, but I can give you an example: file1000685 and file1000568 123wanglijuan @.

> 于2021年12月5日周日 10:48写道： … <#m_9117533523832606587_> Hi, you can read the keys in the dictionary, and select the paired data according the keys 123wanglijuan @.> 于2021年12月3日周五 08:18写道： … <#m_3085210963525479724_> Hi friends,how can select the paired data?What are the characteristics of the paired data — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 https://github.com/chunmeifeng/MTrans/issues/1 (comment) <#1 (comment) https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-985273392>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN2G2MO27T7S7XGMT7DUPBVKPANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub . Thank you very much,I read attributes of the HDF5 files, which can select the paired data according the keys — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment) https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-986198192>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN6CSMAWPRGJVOXQJNDUPMYM7ANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub .

Thank you very much, I have selected the paired data based on your method.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-986221985, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN2JA5DJKW7LVZ22PODUPNLRVANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[123wanglijuan]

Thanks!

[123wanglijuan]

hi, in this file 'MTrans-main\data\math.py' line 66,87,
"data = torch.fft(data, 2, normalized=True)", "data = torch.ifft(data, 2, normalized=True)"; torch has no attribute 'fft','ifft', should be " torch.fft.fft(), torch.fft.fft(t) " so what I should do

[chunmeifeng]

Hi, I think you need to choose a right torch version

123wanglijuan @.***> 于2021年12月12日周日 04:40写道：

hi, in this file 'MTrans-main\data\math.py' line 66,87, "data = torch.fft(data, 2, normalized=True)", "data = torch.ifft(data, 2, normalized=True)"; torch has no attribute 'fft','ifft', should be " torch.fft.fft(), torch.fft.fft(t) " so what I should do

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-991829460, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XNZ53CBA7M6ULCDL47DUQQKUHANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[Nancy0220]

Hi friends,how can select the paired data?What are the characteristics of the paired data

hi friend，I still don't understand how to select the paired date and get the csv file, can you share your csv file or the code for data processing? Thanks!!!

[chunmeifeng]

Hi friends, I don't remember the details, cause I have finished my internship at IIAI, I cannot access the data on the cluster now.

Here, I can give you an example: file1000685 and file1000568. You can read the dictionary information of these two files to view.

Best

Nancy0220 @.***> 于2021年12月20日周一 10:25写道：

Hi friends,how can select the paired data?What are the characteristics of the paired data

hi friend，I still don't understand how to select the paired date and get the csv file, can you share your csv file or the code for data processing? Thanks!!!

— Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-997747151, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN7Y7WKN2CIYU7AE6YDUR3ZBLANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you commented.Message ID: @.***>

[chunmeifeng]

I will select the pair-data list and update the Github page as soon as possible

Nancy0220 @.***> 于2021年12月20日周一 10:25写道：

Hi friends,how can select the paired data?What are the characteristics of the paired data

hi friend，I still don't understand how to select the paired date and get the csv file, can you share your csv file or the code for data processing? Thanks!!!

— Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-997747151, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN7Y7WKN2CIYU7AE6YDUR3ZBLANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you commented.Message ID: @.***>

[123wanglijuan]

Hi friends,how can select the paired data?What are the characteristics of the paired data

hi friend，I still don't understand how to select the paired date and get the csv file, can you share your csv file or the code for data processing? Thanks!!!

with h5py.File(os.path.join(path, file_name), 'r') as f:
    attrs = dict(f.attrs)
    print(attrs)

you can see attrs['acquisition'] = 'CORPDFS_FBK' or 'CORPD_FBK'

[123wanglijuan]

Hi, I think you need to choose a right torch version 123wanglijuan @.***> 于2021年12月12日周日 04:40写道： … hi, in this file 'MTrans-main\data\math.py' line 66,87, "data = torch.fft(data, 2, normalized=True)", "data = torch.ifft(data, 2, normalized=True)"; torch has no attribute 'fft','ifft', should be " torch.fft.fft(), torch.fft.fft(t) " so what I should do — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XNZ53CBA7M6ULCDL47DUQQKUHANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

thank you, I can run successful now

[chunmeifeng]

Hi friends, I don't remember the details, cause I have finished my internship at IIAI, I cannot access the data on the cluster now.

Here, I can give you a paired example: file1000685 and file1000568. You can read the dictionary information of these two files to view.

I will re-filter out the paired data this week, and upload the csv.file as soon as possible.

Best

123wanglijuan @.***> 于2021年12月20日周一 13:18写道：

Hi friends,how can select the paired data?What are the characteristics of the paired data

hi friend，I still don't understand how to select the paired date and get the csv file, can you share your csv file or the code for data processing? Thanks!!!

with h5py.File(os.path.join(path, file_name), 'r') as f:

attrs = dict(f.attrs)

print(attrs)

you can see attrs['acquisition'] = 'CORPDFS_FBK' or 'CORPD_FBK'

— Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-997873788, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN5TI77PZDXXCYSO3TDUR4NJVANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you commented.Message ID: @.***>

[chunmeifeng]

Hi all, I have uploaded the csv.files of the paired data!

[chunmeifeng]

Hi all, I have uploaded the csv.files of the paired data!

123wanglijuan @.***> 于2021年12月20日周一 13:20写道：

Hi, I think you need to choose a right torch version 123wanglijuan @.***> 于2021年12月12日周日 04:40写道： … <#m_-613664193071309936_> hi, in this file 'MTrans-main\data\math.py' line 66,87, "data = torch.fft(data, 2, normalized=True)", "data = torch.ifft(data, 2, normalized=True)"; torch has no attribute 'fft','ifft', should be " torch.fft.fft(), torch.fft.fft(t) " so what I should do — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment) https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-991829460>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XNZ53CBA7M6ULCDL47DUQQKUHANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub .

thank you, I can run successful now

— Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-997875602, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN2MYUVB4BZZVDLXQBTUR4NSRANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you commented.Message ID: @.***>

[chunmeifeng]

Hi all, I have uploaded the csv.files of the paired data!

Y445n9 @.***> 于2021年10月29日周五 11:33写道：

same problem, we have no csv file ,I wonder if author can provide this file?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-954595881, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN3TA4P2RIHGYEDUCK3UJJS6RANCNFSM5ETBTNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[chunmeifeng]

很多人跑了这个数据集，都没有遇到这个问题

发自我的iPad

在 2022年6月21日，10:56，chenjiachengzzz @.***> 写道：

 您好，请问你们有出现这个问题吗，OSError: Unable to open file (unable to open file: name = '/home/a6000/cjc/dataset/mri_data/fast_mri/knee_singlecoil/singlecoil_val/file1002538.h5', errno = 2, error message = 'No such file or directory', flags = 0, o_flags = 0)， 我使用的数据集是fastmri knee_singleceil，但是，我发现我的val文件夹中，没有这个文件。。。 你们下载下来的数据里都有这个文件吗？

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.

[huihui43]

Excellent work! BTW, is the provided CSV file the same as the one used in the original paper? Thx!

[chunmeifeng]

yes

2022年11月21日 下午2:08，寂寞的是烟花 @.***> 写道：

Excellent work! BTW, is the provided CSV file the same as the one used in the original paper? Thx!

— Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/MTrans/issues/1#issuecomment-1321506853, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XNZYFBP7R3OXGNJXI7TWJMGVZANCNFSM5ETBTNBA. You are receiving this because you commented.

[huihui43]

thanks!