-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
ADD downstream and intrinsic evaluation
- Loading branch information
Damianos Melidis
committed
Apr 22, 2020
1 parent
18b9eb2
commit 91635a2
Showing
610 changed files
with
487,147 additions
and
0 deletions.
There are no files selected for viewing
14,435 changes: 14,435 additions & 0 deletions
14,435
downstream_evaluation/new/data/ec/new_dataset.csv
Large diffs are not rendered by default.
Oops, something went wrong.
22,169 changes: 22,169 additions & 0 deletions
22,169
downstream_evaluation/new/data/ec/new_dataset_final.csv
Large diffs are not rendered by default.
Oops, something went wrong.
132 changes: 132 additions & 0 deletions
132
downstream_evaluation/new/data/ec/new_dataset_removed.csv
Large diffs are not rendered by default.
Oops, something went wrong.
4,332 changes: 4,332 additions & 0 deletions
4,332
downstream_evaluation/new/data/ec/new_dataset_test.csv
Large diffs are not rendered by default.
Oops, something went wrong.
10,104 changes: 10,104 additions & 0 deletions
10,104
downstream_evaluation/new/data/ec/new_dataset_train.csv
Large diffs are not rendered by default.
Oops, something went wrong.
14,435 changes: 14,435 additions & 0 deletions
14,435
downstream_evaluation/new/data/ec/new_dataset_train_test.csv
Large diffs are not rendered by default.
Oops, something went wrong.
Binary file added
BIN
+206 KB
...stream_evaluation/new/data/ec/oov_stats/new_dataset_common_oov_domains_hist.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added
BIN
+176 KB
...tream_evaluation/new/data/ec/oov_stats/new_dataset_test_oov_percentage_hist.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,307 @@ | ||
| /home/damian/anaconda3/envs/gensim/bin/python /home/damian/Documents/L3S/projects/dom2vec/code/extrinsic_eval_run.py | ||
| === NEW Experiment === | ||
| 2) data set -> shuffled data set. | ||
| Loading csv(s) -> data set. | ||
| Using columns ['id', 'ec', 'seq', 'interpro_domains'] | ||
| Current df shape: (22168, 4) | ||
| Data set shape: (22168, 4) | ||
| Cleaning rows of csv(s) from no Interpro domains. | ||
| Using columns ['id', 'ec', 'seq', 'interpro_domains'] | ||
| Removing duplicate instances joining x:interpro_domains and y:ec columns | ||
| Data set shape: (14565, 4) | ||
| Count proteins per domain type: | ||
| Protein instances with ONLY unknown domains: 0.8444902162718846% | ||
| Protein instances with ONLY gap domains: 0.05492619292825266% | ||
| Protein instances with ONLY gap and unknown domains: 0.0% | ||
| Protein instances with ONLY Interpro domains: 99.10058359079986% | ||
| Removing protein instances with only full length unknown or only GAP domains | ||
| Starting data set shape: (14565, 4) | ||
| /home/damian/Documents/L3S/projects/dom2vec/code/PrepareDataSet.py:447: UserWarning: Boolean Series key will be reindexed to match DataFrame index. | ||
| self.dataset.drop(self.dataset[one_uniq_domain & gap_domain].index, inplace=True) | ||
| Removed instances data frame shape: (131, 4) | ||
| Final data set data frame shape: (14434, 4) | ||
| Shuffling data set.. | ||
| Saving dataset -> csv. | ||
| ===== | ||
| 3.1) split train/test stratified on label | ||
| Performing train/test split stratified for ec, test size: 0.3. | ||
| Writing: | ||
| dataset_train: (10103, 4) | ||
| dataset_test: (4331, 4) | ||
| --- | ||
| ===== | ||
| 3.2) Check domains distribution in train and test | ||
| Plot common and OOV domains occurrence histograms in test set | ||
| === Stats === | ||
| num of train unique domains: 3219 | ||
| num of test unique domains: 969 | ||
| num of common unique domains: 4549 | ||
| === === === | ||
| Plot OOV percentage per protein in test set | ||
| ===== | ||
| 3.3) create inner 3-fold cross validation sets | ||
| Splitting training set with inner 3-fold stratified on ec. | ||
| Writing fold 1 | ||
| train data: (6735, 4) | ||
| validation data: (3368, 4) | ||
| Writing fold 2 | ||
| train data: (6735, 4) | ||
| validation data: (3368, 4) | ||
| Writing fold 3 | ||
| train data: (6736, 4) | ||
| validation data: (3367, 4) | ||
| --- | ||
| ===== | ||
| 3.4) subsample train set | ||
| Subseting training set in 20 realizations for each percentage in [0.01, 0.02, 0.05, 0.1, 0.2, 0.5] | ||
| --- | ||
| 0 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 1 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 2 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 3 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 4 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 5 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 6 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 7 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 8 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 9 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 10 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 11 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 12 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 13 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 14 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 15 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 16 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 17 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 18 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| 19 instance of percentage 0.01 | ||
| subset of train: (101, 4) | ||
| --- | ||
| 0 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 1 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 2 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 3 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 4 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 5 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 6 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 7 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 8 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 9 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 10 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 11 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 12 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 13 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 14 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 15 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 16 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 17 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 18 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| 19 instance of percentage 0.02 | ||
| subset of train: (202, 4) | ||
| --- | ||
| 0 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 1 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 2 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 3 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 4 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 5 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 6 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 7 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 8 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 9 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 10 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 11 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 12 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 13 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 14 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 15 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 16 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 17 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 18 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| 19 instance of percentage 0.05 | ||
| subset of train: (505, 4) | ||
| --- | ||
| 0 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 1 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 2 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 3 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 4 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 5 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 6 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 7 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 8 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 9 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 10 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 11 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 12 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 13 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 14 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 15 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 16 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 17 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 18 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| 19 instance of percentage 0.1 | ||
| subset of train: (1010, 4) | ||
| --- | ||
| 0 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 1 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 2 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 3 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 4 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 5 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 6 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 7 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 8 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 9 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 10 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 11 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 12 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 13 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 14 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 15 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 16 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 17 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 18 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| 19 instance of percentage 0.2 | ||
| subset of train: (2021, 4) | ||
| --- | ||
| 0 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 1 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 2 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 3 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 4 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 5 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 6 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 7 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 8 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 9 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 10 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 11 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 12 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 13 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 14 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 15 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 16 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 17 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 18 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| 19 instance of percentage 0.5 | ||
| subset of train: (5052, 4) | ||
| === | ||
| === === | ||
|
|
||
| Process finished with exit code 0 | ||
|
|
Oops, something went wrong.