Fix directory links (blob->tree) so folder navigation works; dedupe repeated thesis logo

README.md

@@ -147,7 +147,7 @@ Full tables: [model comparison](tables/model_comparison.csv), [six-variation swe
 
 SHAP analysis on GloVe+BiLSTM shows the model attends to profanity and abuse tokens for the hate class.
 False negatives concentrate on obfuscated or romanized Hindi abuse and code-switch boundaries; false
-positives on aggressive but non-hateful phrasing. Plots: [`phase5/shap/`](phase5/shap).
+positives on aggressive but non-hateful phrasing. Plots: [`phase5/shap/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase5/shap).
 
 ## 5. Discussion
 

THESIS.md

@@ -48,7 +48,6 @@
 
 > **Session: 2022-2026**
 >
-> <img src="thesis_media/media/image1.jpeg" style="width:3.32826in;height:0.94444in" alt="A logo for a university Description automatically generated" />
 
 # 
 
@@ -68,7 +67,6 @@
 >
 > **Place: Guwahati**
 >
-> <img src="thesis_media/media/image1.jpeg" style="width:3.28438in;height:1.075in" alt="A logo for a university Description automatically generated" />
 
 #  FORWARDING CERTIFICATE
 

phase1/README.md

@@ -20,5 +20,5 @@
   lookups, setting the baseline for the hybrid models.
 
 ## Files
-- Notebooks: [`notebooks/`](notebooks/) - Dataset_Split, Fasttext
-- Figures: [`figures/`](figures/) - accuracy/loss curves, confusion matrix, ROC
+- Notebooks: [`notebooks/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase1/notebooks) - Dataset_Split, Fasttext
+- Figures: [`figures/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase1/figures) - accuracy/loss curves, confusion matrix, ROC

phase2/README.md

@@ -18,6 +18,6 @@
   strategy in Phase 3.
 
 ## Files
-- Notebooks: [`notebooks/`](notebooks/) - fasttext_bilstm_monolingual
-- Figures: [`figures/`](figures/) - confusion matrix, ROC, train/val accuracy and loss
+- Notebooks: [`notebooks/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase2/notebooks) - fasttext_bilstm_monolingual
+- Figures: [`figures/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase2/figures) - confusion matrix, ROC, train/val accuracy and loss
 - Tables: [`tables/metrics_table.csv`](tables/metrics_table.csv)

phase3/README.md

@@ -24,6 +24,6 @@
   motivating the multi-stage curriculum in Phase 6.
 
 ## Files
-- Figures: [`figures/`](figures/) - per-language accuracy curves
-- Tables: [`tables/`](tables/) - strategy_results_table, master_metrics, final_results
-- Models: [`models/`](models/) - english_bilstm.h5, hindi_bilstm.h5, hinglish_bilstm.h5
+- Figures: [`figures/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase3/figures) - per-language accuracy curves
+- Tables: [`tables/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase3/tables) - strategy_results_table, master_metrics, final_results
+- Models: [`models/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase3/models) - english_bilstm.h5, hindi_bilstm.h5, hinglish_bilstm.h5

phase4/README.md

@@ -24,7 +24,7 @@
 - High specificity but low recall: the LSTM variants lean toward the majority (non-hate) class.
 
 ## Files
-- Notebooks: [`notebooks/`](notebooks/) - many_to_one_lstm, ManytooneLSTM
-- Figures: [`figures/`](figures/) - per-language accuracy curves
-- Tables: [`tables/`](tables/) - master_metrics, metrics
-- Models: [`models/`](models/) - combined/english/hindi/hinglish BiLSTM
+- Notebooks: [`notebooks/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase4/notebooks) - many_to_one_lstm, ManytooneLSTM
+- Figures: [`figures/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase4/figures) - per-language accuracy curves
+- Tables: [`tables/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase4/tables) - master_metrics, metrics
+- Models: [`models/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase4/models) - combined/english/hindi/hinglish BiLSTM

phase5/README.md

@@ -16,8 +16,8 @@
   obfuscated or romanized Hindi abuse, a concrete target for future preprocessing.
 
 ## Files
-- Figures: [`figures/`](figures/) - per-language confusion matrix, ROC, PR, F1, train curves
-- SHAP: [`shap/`](shap/) - false-positive and false-negative force plots plus waterfall samples
+- Figures: [`figures/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase5/figures) - per-language confusion matrix, ROC, PR, F1, train curves
+- SHAP: [`shap/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase5/shap) - false-positive and false-negative force plots plus waterfall samples
   (representative subset; the full set of ~1,390 plots was generated during training)
-- Tables: [`tables/`](tables/) - final_results, fp_fn_samples
-- Models: [`models/`](models/) - english/hindi/hinglish/full_dataset best (.h5)
+- Tables: [`tables/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase5/tables) - final_results, fp_fn_samples
+- Models: [`models/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase5/models) - english/hindi/hinglish/full_dataset best (.h5)

phase6/README.md

@@ -29,10 +29,10 @@
   minority-language failure mode rather than a working result.
 
 ## Files
-- Notebooks: [`notebooks/`](notebooks/) - regular and sequential training for each embedding,
+- Notebooks: [`notebooks/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase6/notebooks) - regular and sequential training for each embedding,
   plus glove_bilstm_shap_xai
-- Figures: [`figures/`](figures/) - confusion / ROC / accuracy / loss for all six orderings
-- Tables: [`tables/`](tables/) - per-ordering result tables and final_results.
+- Figures: [`figures/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase6/figures) - confusion / ROC / accuracy / loss for all six orderings
+- Tables: [`tables/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase6/tables) - per-ordering result tables and final_results.
   Note: the per-ordering CSVs are the raw training exports (headerless). A clean, labeled
   version is at [`../tables/multistage_six_variations_glove_bilstm.csv`](../tables/multistage_six_variations_glove_bilstm.csv).
-- Models: [`models/`](models/) - one model per language ordering (.h5)
+- Models: [`models/`](https://huggingface.co/pankajbiswas6/hinglish-hate-speech-bilstm/tree/main/phase6/models) - one model per language ordering (.h5)