diff --git a/.gitignore b/.gitignore index ac5aa4b67363e07a02541935c18acf2a302b4141..72fcf7413a2fe552e69233366f5ce86070330bf2 100644 --- a/.gitignore +++ b/.gitignore @@ -1,4 +1,4 @@ -deepgrail_Tagger +SuperTagger Utils/silver Utils/gold .idea diff --git a/Linker/Linker.py b/Linker/Linker.py index 816b5893e45c3eddae25f9b99601482ab0a1b909..ff37ef83fdeec183faf6c6e24fb110c552de9d43 100644 --- a/Linker/Linker.py +++ b/Linker/Linker.py @@ -6,7 +6,6 @@ import time import torch import torch.nn.functional as F -from torch import Module from torch.nn import Sequential, LayerNorm, Dropout from torch.optim import AdamW from torch.utils.data import TensorDataset, random_split @@ -23,7 +22,7 @@ from Linker.atom_map import atom_map from Linker.eval import mesure_accuracy, SinkhornLoss from Linker.utils_linker import find_pos_neg_idexes, get_atoms_batch, FFN, get_axiom_links, get_pos_encoding_for_s_idx, \ get_neg_encoding_for_s_idx -from Supertagger import * +from SuperTagger import * from utils import pad_sequence def format_time(elapsed): diff --git a/train.py b/train.py index 4c6645ea9b8e93faa610e8e10a9b6387f7ea1f02..50354a06864fa0665d0b63927af58c976c090dd8 100644 --- a/train.py +++ b/train.py @@ -5,13 +5,13 @@ from utils import read_csv_pgbar torch.cuda.empty_cache() batch_size = int(Configuration.modelTrainingConfig['batch_size']) -nb_sentences = batch_size * 40 +nb_sentences = batch_size * 2 epochs = int(Configuration.modelTrainingConfig['epoch']) file_path_axiom_links = 'Datasets/gold_dataset_links.csv' df_axiom_links = read_csv_pgbar(file_path_axiom_links, nb_sentences) print("Linker") -linker = Linker("models/model_supertagger.pt") -print("Linker Training") +linker = Linker("models/flaubert_super_98%_V2_50e.pt") +print("\nLinker Training\n\n") linker.train_linker(df_axiom_links, validation_rate=0.1, epochs=epochs, batch_size=batch_size, checkpoint=False, tensorboard=True)