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.

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

Dev pnn pr #340

Open
wants to merge 9 commits into
base: main
Choose a base branch
from
Open

Dev pnn pr #340

wants to merge 9 commits into from

Conversation

BakerMara
Copy link
Contributor

No description provided.

@BakerMara BakerMara requested a review from ShawnXuan May 20, 2022 10:24
pnn_train_eval.py \
--data_dir $DATA_DIR \
--persistent_path $PERSISTENT_PATH \
--table_size_array "43, 98, 121, 41, 219, 112, 79, 68, 91, 5, 26, 36, 70, 1447, 554, 157461, 117683, 305, 17, 11878, 629, 4, 39504, 5128, 156729, 3175, 27, 11070, 149083, 11, 4542, 1996, 4, 154737, 17, 16, 52989, 81, 40882" \
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

replace with new list

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

pnn_train_eval.py \
--data_dir $DATA_DIR \
--persistent_path $PERSISTENT_PATH \
--table_size_array "43, 98, 121, 41, 219, 112, 79, 68, 91, 5, 26, 36, 70, 1447, 554, 157461, 117683, 305, 17, 11878, 629, 4, 39504, 5128, 156729, 3175, 27, 11070, 149083, 11, 4542, 1996, 4, 154737, 17, 16, 52989, 81, 40882" \
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

replace with new list

return self.one_embedding.forward(ids)


class DenseLayer(nn.Module):
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ueless

one_embedding_store_type="cached_host_mem",
cache_memory_budget_mb=8192,
interaction_type="dot",
interaction_itself=False,
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

uesless

self.input_dim = embedding_vec_size * self.fields
if self.use_inner:
self.input_dim += sum(range(self.fields))
self.inner_product_layer = InnerProductLayer(
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copy link
Contributor

@ShawnXuan ShawnXuan left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

OutterProductLayer not work!!!

I = self.inner_product_layer(E)
if self.use_outter:
O = self.outter_product_layer(
E.reshape(self.fields, -1, 1, self.embedding_vec_size)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

what is the purpose of reshape here???

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

E的形状本来是(b, fields, embedding_vec_size), reshape之后是(fields, b, 1, embedding_vec_size)。为了传入opnn网络进行训练。

Copy link
Contributor

@ShawnXuan ShawnXuan left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

suggest to keep IPNN only currently.

@Ldpe2G Ldpe2G mentioned this pull request Jul 7, 2022
11 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants