Social_Media/pytorch-sentiment-analysis
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Commit Graph

  • c0311f4057 updated nbow notebook to end of data preparation refactor bentrevett 2022-07-02 19:19:14 +0100
  • 3365c6a700
         Merge 97f621a8ca into b4efbefa47 Peng Wang 2022-05-05 23:09:36 +0800
  • 97f621a8ca
         Update 6 - Transformers for Sentiment Analysis.ipynb Peng Wang 2022-05-05 23:08:17 +0800
  • daa9028a41 refactored dataloader bentrevett 2022-05-03 22:24:20 +0100
  • 5d8d504d19
         Merge 08d8e69d6c into b4efbefa47 Jay Liu 2021-07-30 08:28:57 -0700
  • b4efbefa47 started writing nbow model text master bentrevett 2021-07-15 22:41:50 +0100
  • 5c2ac07720 added transformers notebook bentrevett 2021-07-08 19:05:29 +0100
  • ef687f9e92 added tqdm bentrevett 2021-07-08 19:05:18 +0100
  • e8f7d1ef96 remove unnecessary semicolons bentrevett 2021-07-08 15:14:05 +0100
  • a610a6cf0b add cnn model bentrevett 2021-07-08 15:13:04 +0100
  • bd32e23e1d add plots to lstm model and change hyperparams bentrevett 2021-07-08 15:12:30 +0100
  • 838cc46826 add plots to nbow model bentrevett 2021-07-08 15:11:54 +0100
  • c8d4c1d136 fix lstm notebook - added missing pad_index to embedding layer - increased dropout bentrevett 2021-07-08 11:43:28 +0100
  • bc5e0f492a cleanup dataset.map functions bentrevett 2021-07-08 11:26:28 +0100
  • 0b62a21dc0 added lstm with torchtext 0.10 bentrevett 2021-07-08 11:12:50 +0100
  • b7debfc0a2 added parameter count - added function to count parameters - small variable name change, text -> ids, and labels -> label bentrevett 2021-07-08 11:12:28 +0100
  • d2e7c9a2a3 fix max_length in 1_nbow - actually use max_length arg when getting tokens from text bentrevett 2021-07-07 18:30:46 +0100
  • 3f9eedb4d4 added notebook for torchtext 0.10 bentrevett 2021-07-07 18:04:25 +0100
  • bc5b2e0221 added torchtext 0.10 imdb on nbow torchtext09 bentrevett 2021-07-07 14:43:27 +0100
  • 2b666b3cba update to torchtext 0.9 bentrevett 2021-03-12 13:25:47 +0000
  • ea3253ddee Merge remote-tracking branch 'origin/master' into torchtext08 torchtext08 bentrevett 2021-03-12 13:18:47 +0000
  • d925186eb4 Merge branch 'master' of https://github.com/bentrevett/pytorch-sentiment-analysis bentrevett 2021-03-12 13:16:24 +0000
  • 6b43b7ca90 lint readme bentrevett 2021-03-12 13:04:25 +0000
  • 7fca96f700 update transformers version in the readme bentrevett 2021-03-12 13:02:28 +0000
  • 12646431b9 update readme bentrevett 2021-03-12 13:01:37 +0000
  • 1192ef8ecf update torchtext08 branch bentrevett 2021-03-12 12:45:14 +0000
  • ba82b9231e updated tutorials + readme with latest versions of libs. bentrevett 2021-02-17 12:52:57 +0000
  • 3cea8e83b8 updated readme for experimental requirements bentrevett 2020-11-16 21:50:28 +0000
  • 69fc4b0c69 refactor of data loading - ignore trying to keep things as generators bentrevett 2020-10-13 21:54:44 +0100
  • db59992e69 updated experimental transformer notebook - now construct each element from "scratch" - view attention averaged over each head - show why problematic example is always negative bentrevett 2020-10-13 17:12:12 +0100
  • 9a5f417d06 added nbow with embeddingbag example bentrevett 2020-09-14 22:31:16 +0100
  • f128236623 updated readme to mention experimental must be run from torchtext master. bentrevett 2020-09-13 22:12:20 +0100
  • b15b88ae50 fix bug with max_length in tokenizer. improved loading vectors. added weight initialization bentrevett 2020-09-13 22:10:45 +0100
  • 42e9e4850d updated transformer with init bentrevett 2020-09-08 23:04:30 +0100
  • 2e9c4f7f9d updates bentrevett 2020-09-08 16:28:04 +0100
  • d9520e6aad added experimental cnn and transformer bentrevett 2020-08-27 17:31:40 +0100
  • 270c1e16c1 updated experimental bentrevett 2020-08-25 18:15:24 +0100
  • c0a31c6b17 updated experimental Ben Trevett 2020-08-18 22:37:09 +0100
  • ac71dbd87d fixed experimental folder link. bentrevett 2020-08-10 11:45:45 +0100
  • d88f3c6deb
         Merge pull request #82 from bentrevett/experimental Ben Trevett 2020-08-10 11:43:30 +0100
  • b8ca165eb7 moved experimental to /experimental (for now). updated readme. experimental bentrevett 2020-08-10 11:42:45 +0100
  • e48124f292 added experimental nbow model bentrevett 2020-08-10 11:30:17 +0100
  • d7be1abb54 added experimental nbow model bentrevett 2020-08-03 22:58:48 +0100
  • cac01789be
         Merge pull request #80 from Spiderpig86/master Ben Trevett 2020-07-18 19:45:45 +0100
  • fd40c22a9a
         Fixed broken PyTorch setup link Stanley Lim 2020-07-17 23:46:16 -0700
  • 08d8e69d6c
         install spacy english model Jay Liu 2020-06-12 13:45:39 +1000
  • 79bb86abc9 clarified loading own datasets bentrevett 2020-01-28 09:53:16 +0000
  • 4aa27b0db3 (#63) fixed typo in transformer model bentrevett 2020-01-10 13:47:27 +0000
  • 9210842371 updated readme to mention updated pytorch version requirement bentrevett 2019-11-25 14:45:30 +0000
  • 97b909cdef removed commented out imports bentrevett 2019-11-06 16:06:58 +0000
  • 086d9ef5a6 updated readme with details about transformer bentrevett 2019-11-06 16:04:40 +0000
  • 2e3ea700d5 renamed transformers tutorial, added initial comments bentrevett 2019-11-06 15:52:50 +0000
  • 492b6d572a added batch_first to remove a permute bentrevett 2019-11-06 14:01:23 +0000
  • 9ae32cf124 added transformer notebook results bentrevett 2019-11-06 00:35:56 +0000
  • 96fe748dac added initial transformers tutorial bentrevett 2019-11-05 23:36:23 +0000
  • 0e8eca1e85 updated gitignore for vs code bentrevett 2019-11-05 23:36:06 +0000
  • 0356b3c035 fixed appendix C loading incorrect embeddings from cache bentrevett 2019-09-24 16:35:57 +0100
  • 67eba4df76 added colab links for all tutorials bentrevett 2019-09-20 11:42:39 +0100
  • a43251b47d testing adding colab badge/links bentrevett 2019-09-20 11:39:59 +0100
  • fe430c603e added explicit notes to copy embeddings using weight.data and not weight bentrevett 2019-09-19 16:32:07 +0100
  • ed9455e89a non-deterministic error fixed in latest cudnn version bentrevett 2019-09-19 16:16:22 +0100
  • 2753e93ca4 reran all notebooks with latest pytorch and torchtext to ensure still working bentrevett 2019-09-16 11:32:18 +0100
  • 5b0af623a9 updated readme to note latest versions of pytorch and torchtext are being used bentrevett 2019-09-16 11:31:54 +0100
  • b1db529eee removed unnecessary squeeze in rnn forward method bentrevett 2019-07-10 13:10:30 +0100
  • 808aaafa6f removed extra print statement added whilst debugging. changed default model back to 2d cnn bentrevett 2019-04-22 14:27:48 +0100
  • 807ec2082a fixed typos in max pool figure and size of tensors after convolutional filters (#35) bentrevett 2019-04-22 13:25:46 +0100
  • 61093d819c removed unnecessary torch.cuda.manual_seed from appendix c notebook bentrevett 2019-04-16 10:10:59 +0100
  • ed89d29f8d added some reference links bentrevett 2019-04-16 10:08:07 +0100
  • a901a727e6 removed testing cell from notebook C, updated gitignore to ignore trained embeddings file bentrevett 2019-04-16 09:57:20 +0100
  • 982083f52a updated pytorch version bentrevett 2019-04-16 09:37:43 +0100
  • b337ea6976 updated readme to note bug in pytorch for tutorial 2 bentrevett 2019-04-10 15:35:24 +0100
  • be122f3a6f added model.eval() in predict sentiment functions (#31) bentrevett 2019-04-10 10:27:20 +0100
  • 0e2e8f49ba updated readme to link to appendix C bentrevett 2019-04-02 13:39:51 +0100
  • faaa6a1adc added appendix c - handling embeddings bentrevett 2019-04-02 13:36:14 +0100
  • 856d9e5598 updated appendix B - formatting and typos bentrevett 2019-04-01 17:08:38 +0100
  • 669a9b8436 removed erroneous statement about the cnn model being the fastest model bentrevett 2019-04-01 16:40:55 +0100
  • 36c8bf7cdc updated readme for some more details on appendix b bentrevett 2019-04-01 16:19:07 +0100
  • 9403847535 updated readme to mention the use of packed padded sequences in notebook 2 bentrevett 2019-04-01 16:18:02 +0100
  • c429c70e73 changed some wording due to rerunning notebook 2 with packed padded sequences bentrevett 2019-04-01 16:11:52 +0100
  • c13c674a67 added packed padded sequences to notebook 2 bentrevett 2019-04-01 16:08:53 +0100
  • 63c08ef0bc mentioned how notebook 2 will introduce packed padded sequences bentrevett 2019-04-01 16:08:30 +0100
  • 525ae0b416 lots of formatting changes bentrevett 2019-03-29 16:57:00 +0000
  • eea44cb487 added missing unk init and fixed some formatting to notebook 4 bentrevett 2019-03-29 16:37:21 +0000
  • b21f42f764 mention how using sort key is preferred over sort = False bentrevett 2019-03-23 14:48:52 +0000
  • 59920c56cf removed incorrect xavier init example from notebook 2 bentrevett 2019-03-23 14:45:52 +0000
  • b9d46f0ce0 added need for sort/sort_key in iterators when using own dataset (#27) bentrevett 2019-03-23 10:06:34 +0000
  • bf8cc46e48 added parameter count and epoch timer functions to all notebooks. also added cnn1d imlementation to notebook 4 bentrevett 2019-03-21 22:48:21 +0000
  • cd09d20b65 fixed out of glove vector initialization and missing generate bigrams functions in fasttext implementation bentrevett 2019-03-21 17:22:25 +0000
  • 5e7c5a9fcf added text to multi-class notebook bentrevett 2019-03-10 17:30:07 +0000
  • 6e037d8386 fixed wording and updated assets for notebook 4 bentrevett 2019-03-10 16:28:21 +0000
  • ce7ce974c0 updated readme to mention multiclass notebook bentrevett 2019-03-10 15:46:11 +0000
  • 977613215e updated assets and wording for notebook 3 bentrevett 2019-03-10 15:45:40 +0000
  • 14edc64ff8 fixed wording on notebook 2 bentrevett 2019-03-10 15:37:09 +0000
  • 25efef84d4 changed imgur urls to own assets bentrevett 2019-03-10 14:45:44 +0000
  • 0327a54c63 fixed some wording bentrevett 2019-03-10 14:16:18 +0000
  • 61ee0e66a5 Merge branch 'master' of https://github.com/bentrevett/pytorch-sentiment-analysis bentrevett 2019-01-25 14:00:11 +0000
  • 8220b98d2a added rough draft of multi-class notebook bentrevett 2019-01-25 13:56:46 +0000
  • 7afa650a34
         Merge pull request #13 from zdepablo/master Ben Trevett 2018-12-10 23:19:02 +0000
  • 5571f93cb8
         Remove an incomplete phrase (#1) César de Pablo 2018-12-08 22:16:55 +0100
  • 7dd3c19876 fixed typo bentrevett 2018-11-22 21:15:45 +0000