updating with new data and eval results

README.md

@@ -79,6 +79,9 @@ It has been training on
 - [Comm2 - AAC Text](https://www.aactext.org/comm2/)
 - [C4-200M - 25K Subset](https://huggingface.co/datasets/leslyarun/c4_200m_gec_train100k_test25k)
 - [JFLEG](https://huggingface.co/datasets/jhu-clsp/jfleg)
+- [Coedit](https://huggingface.co/datasets/grammarly/coedit)
+- [Conversation Enders](https://huggingface.co/Chakshu/conversation_ender)
+- [Conversation Starters](https://huggingface.co/Langame/conversation-starters)
 
 
 Then injecting  typos from a range of places
@@ -88,6 +91,7 @@ Then injecting  typos from a range of places
 - **TOEFL Spell** A dataset of Spelling Annotations for English language learner essays written for TOEFL exams.
   - Find this [here](https://github.com/EducationalTestingService/TOEFL-Spell/tree/master) 
 - **Homonyms** We replace words in BNC and Dialy Dialog occasionally with homonyms from this list https://github.com/pimentel/homophones/
+- **Our own typo augment function** This would make likely errors found in a English Qwerty layout as well as subsitutions, deletions etc
 
 And then compressing versions of the sentences (i.e. removing spaces)- both correct and typod we add to our dataset. (This is to solve a problem where some people write without spaces)
 Note we use a ``grammar: `` prefix for each sentence in training. 
@@ -113,7 +117,7 @@ Full script to build the [dataset is here](https://colab.research.google.com/dri
 - The model is fine-tuned from `t5-small`.
 
 ## Resources for more information:
-- [GitHub Repo](https://github.com/willwade/dailyDialogCorrections/)
+- [GitHub Repo](https://github.com/AceCentre/Correct-A-Sentence/tree/main/helper-scripts/)
 
 # Uses
 
@@ -141,10 +145,23 @@ The model was trained on a curated subset of the DailyDialog and BNC corpora (20
 Sentences were stripped of apostrophes and commas, spaces were removed, and typos were introduced programmatically to simulate common errors in user-generated content.
 
 ### Speeds, Sizes, Times
-- Training was conducted on LlambdaLabs, taking approximately 4 hrs to complete.
+- Training was conducted on LlambdaLabs, taking approximately 6 hrs to complete.
 
 # Evaluation
 
+| Phase      | Metric                          | Value         |
+|------------|----------------------------------|---------------|
+| Train      | Loss                            | 0.1642        |
+| Train      | Global Step                      | 375876        |
+| Train      | Total FLoPs                     | 8.33E+15      |
+| Train      | Training Time                    | ~6.5 hr       |
+| Eval       | Loss                            | 0.1199        |
+| Eval       | Samples per Second               | 1159.375      |
+| Eval       | Steps per Second                 | 72.462       |
+| Hyperparam | Learning Rate                    | 5.02E-06      |
+| Hyperparam | Grad Norm                       | 1.1725        |
+| Hyperparam | Epoch                           | 3             |
+
 ## Testing Data, Factors & Metrics
 
 ### Testing Data
@@ -156,59 +173,91 @@ The model demonstrates high efficacy in correcting short, erroneous sentences, w
 
 It performs nearly on par with GPTTurbo16k at around 93% sentence similarity. But there are gaps. 
 
-Take for example this output and I've bolded elements for parts that are I feel are incorrect. 
+Take for example this output.  
 
 
+Original: Howwasyafternoonanyway?
+Corrected: How was my afternoon anyway?
 Original: Didyoucatchthegamelastnight?
 Corrected: Did you catch the game last night?
-
 Original: Wannagrabcoffeetomorrow?
 Corrected: Wanna grab coffee tomorrow?
-
 Original: ImdyingsomeonecancellsoIcandogsitter!
-Corrected: I'm dying someone **cancell** so I can dogsitter!
-
+Corrected: I'm dying someone cancell so I can do dogsitter!
 Original: Hahahahahahahathats hilarious!
-Corrected: Haha ha ha ha that's hilarious!
-
+Corrected: Hahahahahahahahaha that's hilarious!
 Original: OMGyouneedtoseethelatestmeme!
-Corrected: OMG you need to see the latest meme!
-
+Corrected: OMG, you need to see the latest me!
 Original: Seriouslythisweatherissocrazy!
-Corrected: Seriously this weather is so crazy!
-
+Corrected: Seriously, this weather is so crazy!
 Original: Whatchauptomefriend?
-Corrected: What's **his** friend?
-
+Corrected: What's your friend?
 Original: Feelingburntoutaftettodayhelp!
-Corrected: Feeling burnt out today help!
-
+Corrected: Feeling burnt out aftet today, help!
 Original: Guesswhosingleagain!
 Corrected: Guess who single again!
-
 Original: Youwontyoubelievewhatjusthappened!
-Corrected: You **want** you believe what just happened!
-
+Corrected: You wont you believe what just happened!
 Original: Moviemarathonatmyplacethisweekend?
-Corrected: Movie Marathon at my place this weekend?
-
+Corrected: Movie marathon at my place this weekend?
 Original: Needstudymotivationanyideas?
-Corrected: Need study motivation any ideas?
-
+Corrected: Need study motivation, any ideas?
 Original: Sostressedaboutthispresentation!
 Corrected: So stressed about this presentation!
-
 Original: Finallyfinishedthatbookyourecommended!
-Corrected: Finally finished that book **you're** recommended!
-
+Corrected: Finally finished that book you recommended!
 Original: Anygoodshowsbingeablelately?
-Corrected: Any good shows **biteable** recently?
+Corrected: Any good shows being possible lately?
+Original: Justsawthecraziestthingonthebus!
+Corrected: Just saw the craziest thing on the bus!
+Original: Sendhelpfoodistrappedintheoven!
+Corrected: Send help food is wrapped in the oven!
+Original: Cantwaittoseeyouattheparty!
+Corrected: Cant wait to see you at the party!
+Original: Missyoutonsalreadyletshangsoon!
+Corrected: Miss youtons already let's hang soon!
+Original: CantbelieveImissedthelastbus!
+Corrected: Can't believe I missed the last bus!
+Original: Needanysuggestionsforagoodmovieatnight?
+Corrected: Need any suggestions for a good movie night?
+Original: Feelingproudaccomplishedsomethingbigtoday!
+Corrected: Feeling proud of something big today!
+Original: Wishcouldteleportmyselftothebeachrightnow.
+Corrected: Wish could teleport myself to the beach right now.
+Original: Justsawthecutestaudiofapuppylearningtotalk.
+Corrected: Just saw the cutest audio from puppy learning to talk.
+Original: Excitedtostartafreshnewprojectthistoday.
+Corrected: Excited to start a fresh new project this today.
+Original: Havingtroubledecidingwhichoptionistobetter.
+Corrected: Having trouble deciding which option is to better.
+Original: Finallyfinishedorganizingmyclosetfeelsamazing!
+Corrected: Finally finished organizing my closet feels amazing!
+Original: Learnedsomethingnewtodayitssoneversadtoolate!
+Corrected: Learned something new today it's so wonderful too late!
+Original: Cravingapizzabuttryingtoresisttemptation.
+Corrected: Craving a pizza, but trying to resist temptation.
+Original: Planningweekendgettogetheranyonesinterested?
+Corrected: Planning weekend get together anyone's interested?
+Original: Canwaittousethisnewlyacquiredskillsoon.
+Corrected: Can wait to use this newly acquired skill soon.
+Original: Feelinggratefulforallofthesupportivepeopleinmylife.
+Corrected: Feeling grateful for all of the supportive people in my life.
+Original: Whatshappeningonthelatestseasonofyourfavoriteshow?
+Corrected: What's happening on the latest season of your favorites, though?
+Original: Anyoneelseafraidofthedarkadmitnojudgement.
+Corrected: Anyone else afraid of the dark admit no judgement.
+Original: Justreadafascinatingarticleaboutancientcivilizations.
+Corrected: Just read a fascinating article about ancient civilizations.
+Original: Feelingaccomplishedcrossedofeverythingontmylisttoday.
+Corrected: Feeling accomplished crossed of everything on my list today.
+Original: Strugglingtofindmotivationanyadviceisappreciated.
+Corrected: Struggling to find motivation any advice is appreciated.
+Original: Cantw waittoseeyouagainletsmakesoonplans!
+Corrected: Can't wait to see you again, let's make soon plans!
 
 We hope to build on this by further fine-tuning in time on real corpous of indviduals using AAC
 
 
-#EvalResult(loss=0.8066404461860657)
-
 
 # Technical Specifications