git-market/rasbt-LLMs-from-scratch
1
0
Fork 0
mirror of https://github.com/rasbt/LLMs-from-scratch.git synced 2026-04-11 14:21:41 +08:00
Code Issues Actions 11 Packages Projects Releases Wiki Activity

make batch loss calculatution more efficient

Browse Source
This commit is contained in:
rasbt 2024-03-27 07:11:56 -05:00
parent 3cb5a52a1b
commit 88b2dd780a
5 changed files with 79 additions and 73 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
appendix-D/01_main-chapter-code/previous_chapters.py
View File

@ -259,6 +259,8 @@ def calc_loss_loader(data_loader, model, device, num_batches=None):
total_loss = 0.
if num_batches is None:
num_batches = len(data_loader)
else:
num_batches = min(num_batches, len(data_loader))
for i, (input_batch, target_batch) in enumerate(data_loader):
if i < num_batches:
loss = calc_loss_batch(input_batch, target_batch, model, device)

144
ch05/01_main-chapter-code/ch05.ipynb
View File

File diff suppressed because one or more lines are too long

2
ch05/01_main-chapter-code/train.py
View File

@ -34,6 +34,8 @@ def calc_loss_loader(data_loader, model, device, num_batches=None):
total_loss = 0.
if num_batches is None:
num_batches = len(data_loader)
else:
num_batches = min(num_batches, len(data_loader))
for i, (input_batch, target_batch) in enumerate(data_loader):
if i < num_batches:
loss = calc_loss_batch(input_batch, target_batch, model, device)

2
ch05/03_bonus_pretraining_on_gutenberg/previous_chapters.py
View File

@ -253,6 +253,8 @@ def calc_loss_loader(data_loader, model, device, num_batches=None):
total_loss = 0.
if num_batches is None:
num_batches = len(data_loader)
else:
num_batches = min(num_batches, len(data_loader))
for i, (input_batch, target_batch) in enumerate(data_loader):
if i < num_batches:
loss = calc_loss_batch(input_batch, target_batch, model, device)

2
ch05/05_bonus_hparam_tuning/hparam_search.py
View File

@ -27,6 +27,8 @@ def calc_loss_loader(data_loader, model, device, num_batches=None):
total_loss = 0.
if num_batches is None:
num_batches = len(data_loader)
else:
num_batches = min(num_batches, len(data_loader))
for i, (input_batch, target_batch) in enumerate(data_loader):
if i < num_batches:
loss = calc_loss_batch(input_batch, target_batch, model, device)