PoliBa-DeepLearning/NanoSocrates
2
0
Fork 0
Code Issues 3 Pull Requests 1 Actions Packages Projects Releases Wiki Activity

dev.train #8

Merged
gape_01 merged 50 commits from dev.train into dev 2025-10-17 22:20:14 +02:00
Conversation 0 Commits 50 Files Changed 49 +6
1 changed files with 6 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files
Showing only changes of commit 37a2501a79 - Show all commits

6
Playgrounds/nanosocrates-train-experiment-2.py
View File

@ -78,6 +78,7 @@ TEST_BATCHER = Batch.Batcher(TEST_DATASET_PATH, SENTENCE_LENGTH, TOKENANO, MASKE
# Model # Model
NANOSOCRATES = Transformer.TrainingModel( NANOSOCRATES = Transformer.TrainingModel(
TOKEN_SPACE_SIZE, TOKEN_SPACE_SIZE,
EMBEDDED_SIZE, EMBEDDED_SIZE,
@ -85,6 +86,11 @@ NANOSOCRATES = Transformer.TrainingModel(
ATTENTION_HEADS, ATTENTION_HEADS,
NUMBER_OF_BLOCKS, NUMBER_OF_BLOCKS,
) )
if CHECKPOINT_PATH.is_file():
nanosocrates_dict = torch.load(CHECKPOINT_PATH, weights_only=True)
NANOSOCRATES.load_state_dict(nanosocrates_dict)
_, ENCODER_ONLY, DECODER_ONLY = TUtils.decompose_nano_socrates( _, ENCODER_ONLY, DECODER_ONLY = TUtils.decompose_nano_socrates(
NANOSOCRATES, TOKEN_SPACE_SIZE, EMBEDDED_SIZE NANOSOCRATES, TOKEN_SPACE_SIZE, EMBEDDED_SIZE
) )