Skip to content
This repository was archived by the owner on Jan 21, 2025. It is now read-only.

Pull requests: tensorflow/mesh

33 Open 279 Closed
33 Open 279 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Update attention.py
#394 opened Oct 4, 2023 by sjw8793 Loading…
1
Move convert_to_tensor, convert_to_tensor_v1, convert_to_tensor_v1_with_dispatch, convert_to_tensor_v2_with_dispatch, and convert_to_tensor_v2 into tensor_conversion_registry.
#389 opened Mar 13, 2023 by copybara-service bot Loading…
Adding a new Gradient Estimator for Routing using REINFORCE with a leave-one-out baseline.
#374 opened Mar 16, 2022 by copybara-service bot Loading…
Minor comment fix to refer to the correct argument name. cla: yes
#367 opened Oct 27, 2021 by copybara-service bot Loading…
Fix some example code in readme for einsum operation cla: yes
#365 opened Oct 10, 2021 by baragona Loading…
2
Output raw model outputs during eval cla: yes
#361 opened Sep 22, 2021 by craffel Loading…
Save scores lazily. cla: yes
#359 opened Sep 10, 2021 by copybara-service bot Loading…
MODE models with hetereogeneous expert width cla: no
#342 opened Jul 28, 2021 by copybara-service bot Loading…
1
Option to use mtf.Print to log which tokens are sent to which experts when run on CPU. cla: yes
#329 opened Jun 17, 2021 by copybara-service bot Loading…
Splitting tokens when routing cla: no
#326 opened Jun 9, 2021 by copybara-service bot Loading…
2
Add original AI2 version of c4 v3.0.1, ND3 deduplicated with param = 0.8, and LM1B, Wiki40B, and lm_first_len512 versions of original AI2 C4 and ND3 deduped AI2 C4 for evaluation. cla: yes
#321 opened May 18, 2021 by copybara-service bot Loading…
Add cast preprocessor and add tasks for inference prompts for deduplication project. cla: yes
#320 opened May 12, 2021 by copybara-service bot Loading…
Allow init_from_checkpoint to accept a list of pairs, so as to enable initialization of multiple variables in the graph from the same variable in the checkpoint. cla: yes
#298 opened Mar 10, 2021 by copybara-service bot Loading…
Allow for disabling the automatic save on shutdown. cla: yes
#293 opened Feb 25, 2021 by copybara-service bot Loading…
Add loss functions for multiple-target objectives for distillation. cla: no
#291 opened Feb 9, 2021 by copybara-service bot Loading…
2
Use multiple target objectives for distillation. Also see cl/356382304 cla: no
#290 opened Feb 9, 2021 by copybara-service bot Loading…
2
internal cla: no
#283 opened Jan 29, 2021 by copybara-service bot Loading…
1
Decode Unicode strings in inference mode. cla: no
#281 opened Jan 28, 2021 by copybara-service bot Loading…
1
Fixing model export breakage. cla: no
#259 opened Dec 11, 2020 by copybara-service bot Loading…
1
Adds spectral functions to Mesh TensorFlow cla: yes
#250 opened Nov 26, 2020 by EiffL Loading…
5
Remove unused vocab argument from dataset function calls. cla: yes
#217 opened Oct 22, 2020 by copybara-service bot Loading…
This shouldn't have public changes once the diffbase is submitted. cla: yes
#191 opened Sep 18, 2020 by copybara-service bot Loading…
doc fixes
#180 opened Sep 7, 2020 by Mistobaan Loading…
Add scores for generated text in inference mode cla: yes
#164 opened Aug 20, 2020 by allen-q Loading…
9
Internal change cla: no
#160 opened Aug 17, 2020 by copybara-service bot Loading…
1
ProTip! Follow long discussions with comments:>50.