Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[Feature request] truncated normal initializer(sampler) #2129

Open
ruotianluo opened this issue Jul 17, 2017 · 1 comment
Open

[Feature request] truncated normal initializer(sampler) #2129

ruotianluo opened this issue Jul 17, 2017 · 1 comment
Labels
enhancement Not as big of a feature, but technically not a bug. Should be easy to fix module: initialization Related to weight initialization on operators triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
Projects
Issue Status Issue Categories

Comments

@ruotianluo
Copy link
Contributor

There has been a recent discussion on the discuss about truncated normal initializer link.
I have implemented one on my own, but the code is basically paraphrasing the tensorflow code, which I feel like may cause problem. If you feel good to add this, I can submit a PR once you tell me if it's ok to paraphrase or what's the alternative.
@soumith soumith added this to Uncategorized in Issue Status Aug 23, 2017
@soumith soumith added this to nn / autograd / torch in Issue Categories Aug 31, 2017
@ruotianluo ruotianluo mentioned this issue Mar 8, 2018
Closed
@houseroad houseroad mentioned this issue Sep 16, 2019
Closed
houseroad added a commit to houseroad/pytorch that referenced this issue Sep 17, 2019
@houseroad @facebook-github-bot
Automatic update of fbcode/onnx to 1316afc9f972f81340faa05763e2898f38…
8ac9087 
…bcc3b0 (pytorch#26309)

Summary:
Pull Request resolved: pytorch#26309

Previous import was 95252c2adec185e305e34486c6756ece9aa8f57f

Included changes:
- **[1316afc9](onnx/onnx@1316afc9)**: Update IR doc to clarify initializers are permitted as node inputs (pytorch#2320) <G. Ramalingam>
- **[5e920d0c](onnx/onnx@5e920d0c)**: Avoid uses of special chars (pytorch#2315) <Wei-Sheng Chin>
- **[2fa08b0f](onnx/onnx@2fa08b0f)**: Regenerate ONNX proto and add release date to ver 6 IR (pytorch#2316) <Wei-Sheng Chin>
- **[adf9c7a3](onnx/onnx@adf9c7a3)**: Add description of default type about y_zero_point (pytorch#2110) <Takeshi Watanabe>
- **[ee7072c7](onnx/onnx@ee7072c7)**: Support make_attribute empty string (pytorch#2129) <shjwudp>
- **[f913b6e7](onnx/onnx@f913b6e7)**: More unsqueeze tests (pytorch#2200) <James Allingham>
- **[57b51937](onnx/onnx@57b51937)**: Fix resize shape inference issue in opset10 (pytorch#2294) <Bowen Bao>
- **[d7595f34](onnx/onnx@d7595f34)**: Sequence related ops (pytorch#2249) <Bowen Bao>
- **[599f3da9](onnx/onnx@599f3da9)**: Add helper function update_inputs_outputs_dims to tools (pytorch#2148) <Bowen Bao>
- **[3e6382bc](onnx/onnx@3e6382bc)**: Update documentation about required input output types (pytorch#2310) <G. Ramalingam>
- **[0c765d9b](onnx/onnx@0c765d9b)**: Shape inference for NMS (pytorch#2269) <Hariharan Seshadri>
- **[89266710](onnx/onnx@89266710)**: Fix extra collect_snippets warning (pytorch#2277) (pytorch#2307) <Lutz Roeder>

Test Plan: ci

Reviewed By: hl475

Differential Revision: D17403954

fbshipit-source-id: 8f7cbf77b7e09b73db2015f3e0f436772482b322
facebook-github-bot pushed a commit that referenced this issue Sep 17, 2019
@houseroad @facebook-github-bot
Automatic update of fbcode/onnx to 1316afc9f972f81340faa05763e2898f38…
bebc3d6 
…bcc3b0 (#26309)

Summary:
Pull Request resolved: #26309

Previous import was 95252c2adec185e305e34486c6756ece9aa8f57f

Included changes:
- **[1316afc9](onnx/onnx@1316afc9)**: Update IR doc to clarify initializers are permitted as node inputs (#2320) <G. Ramalingam>
- **[5e920d0c](onnx/onnx@5e920d0c)**: Avoid uses of special chars (#2315) <Wei-Sheng Chin>
- **[2fa08b0f](onnx/onnx@2fa08b0f)**: Regenerate ONNX proto and add release date to ver 6 IR (#2316) <Wei-Sheng Chin>
- **[adf9c7a3](onnx/onnx@adf9c7a3)**: Add description of default type about y_zero_point (#2110) <Takeshi Watanabe>
- **[ee7072c7](onnx/onnx@ee7072c7)**: Support make_attribute empty string (#2129) <shjwudp>
- **[f913b6e7](onnx/onnx@f913b6e7)**: More unsqueeze tests (#2200) <James Allingham>
- **[57b51937](onnx/onnx@57b51937)**: Fix resize shape inference issue in opset10 (#2294) <Bowen Bao>
- **[d7595f34](onnx/onnx@d7595f34)**: Sequence related ops (#2249) <Bowen Bao>
- **[599f3da9](onnx/onnx@599f3da9)**: Add helper function update_inputs_outputs_dims to tools (#2148) <Bowen Bao>
- **[3e6382bc](onnx/onnx@3e6382bc)**: Update documentation about required input output types (#2310) <G. Ramalingam>
- **[0c765d9b](onnx/onnx@0c765d9b)**: Shape inference for NMS (#2269) <Hariharan Seshadri>
- **[89266710](onnx/onnx@89266710)**: Fix extra collect_snippets warning (#2277) (#2307) <Lutz Roeder>

Test Plan: ci

Reviewed By: hl475

Differential Revision: D17403954

fbshipit-source-id: 78a9c3ecf5aa7f7a0ba8ea30286eab61ee903772
@cossio
Copy link

cossio commented Jan 16, 2020 •
edited

I think this issue should be reopened, since the PR was not merged in the end and the truncated normal is still not available in PyTorch. @soumith

Related: #31945

@cossio cossio mentioned this issue Jan 16, 2020
Open
@soumith soumith reopened this Jan 27, 2020
@cossio cossio mentioned this issue Jan 27, 2020
Open
@zhangguanheng66 zhangguanheng66 added module: operators triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module labels Jan 27, 2020
@Enealor Enealor mentioned this issue Jan 30, 2020
Closed
@mruberry mruberry added module: initialization Related to weight initialization on operators enhancement Not as big of a feature, but technically not a bug. Should be easy to fix and removed module: operators (deprecated) labels Oct 6, 2020
@datumbox datumbox mentioned this issue Sep 4, 2021
Closed
16 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement Not as big of a feature, but technically not a bug. Should be easy to fix module: initialization Related to weight initialization on operators triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
Projects
Issue Categories
neural-nets
Issue Status
Uncategorized
Milestone
No milestone
Development

No branches or pull requests
5 participants
@soumith @zhangguanheng66 @cossio @ruotianluo @mruberry