Which version of TF are you using? #1
8000
Comments
|
We were using tf1, with the (deprecated) library slim. The code should work
up to tf 1.14.
We do not plan to upgrade it to tf v2 as both Harm and I are working on
completely different project now.
Note that if you want to do so it should now be super easy! Happy to help
if necessary.
…On Tue, Dec 24, 2019, 9:13 PM leotam ***@***.***> wrote:
I'm encountering
Tensor name "resnet_v1_152/block1/unit_1/bottleneck_v1/conv1/biases" not
found in checkpoint files /home/referit/data/resnet_v1_152.ckpt [[node
save/RestoreV2 (defined at
/home/referit/src/referit/preprocess_data/generic/preprocess_data/extract_img_features.py:32)
]]
which others suggest may be due to TF version naming
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#1?email_source=notifications&email_token=ADF7HBFEX4IQIZCP7MDESC3Q2JUPNA5CNFSM4J7ASF6KYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4ICRL22Q>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADF7HBFACBWKJWE36RLCFOTQ2JUPNANCNFSM4J7ASF6A>
.
|
|
Hm I'm using conda 4.8.0 with TF 1.13.1 and python 3.6.8. It seemed similar to this issue: tensorflow/models#2527 which suggested syncing TF versions and barring that setting argscope. Since it seems your argscope is set, I assumed it was TF version. Am I missing some argscope parameter? Using: |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I'm encountering
Tensor name "resnet_v1_152/block1/unit_1/bottleneck_v1/conv1/biases" not found in checkpoint files /home/referit/data/resnet_v1_152.ckpt [[node save/RestoreV2 (defined at /home/referit/src/referit/preprocess_data/generic/preprocess_data/extract_img_features.py:32) ]]which others suggest may be due to TF version naming
The text was updated successfully, but these errors were encountered: