homer-skipgram:跳过文本模型（word2Vec）的荷Maven本训练
需积分: 12 94 浏览量
2021-05-22
20:54:25
上传
评论
收藏 53.23MB ZIP 举报
homer-skipgram-main.zip （47个子文件） 
homer-skipgram-main data models 
Skipgram_Pytorch_0602_lemmatized.pth 19.34MB embeddings_lemmatized.npy 9.66MB README.md 52B
stopwords.txt 5KB vocabs Homer_word2index_accented.json 781KB Homer_word2index.json 767KB fasttext_vocab.json 584KB Homer_index2word.json 827KB Homer_word_frequencies_accented.json 1.14MB Homer_wordList.csv 661KB Homer_word_frequencies.json 1.14MB raw_data HomerGesamt_cleaned.txt 2.3MB HomerGesamt.txt 2.31MB HomerGesamt_deaccented.txt 4.84MB assets 
losses_plot.png 2.25MB torch_skip_0502_euclidean_tsne_perpl7.png 193KB torch_skip_0502_euclidean_tsne_perpl20.png 166KB
tsne_fasttext.html 1.87MB test_optimizers losses_train_Adadelta.png 3.16MB losses_train_Adamax.png 1.93MB losses_train_sgd_nesterov.png 3.27MB losses_train_Adagrad.png 1.92MB losses_train_RMSop_optimizer.png 759KB Loss_Skipgram_after_3_epochen.png 61KB tsne_plot.html 1.86MB torch_skip_0502.png 191KB README.md 86B torch_skip_0502_euclidean_tsne.png 193KB Homer_tokenized_corpus.npy 8.48MB Homer_skipgram_dataset_accented.npy 7.57MB Homer_skipgram_dataset.npy 20.93MB Homer_tokenized_accented.npy 3.33MB skipgram_homer.py 7KB preprocessing preprocessing.py 4KB LICENSE 1KB skipgram_eval.ipynb 882KB train_skipgram.py 5KB requirements.txt 9KB .gitignore 2KB fasttext visualize.py 2KB homer_fastText.py 1KB README.md 60B README.md 6KB utils utils.py 4KB dataset.py 3KB __init__.py 0B modules.py 5KB资源评论
