Model save

README.md

@@ -7,8 +7,6 @@ tags:
 model-index:
 - name: flashcarder
   results: []
-datasets:
-- dpeelen9/academic-flashcards
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,26 +14,24 @@ should probably proofread and complete it, then remove this comment. -->
 
 # flashcarder
 
-This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on an my own academic-flashcards dataset.
+This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.223
+- Loss: 0.1990
 
 ## Model description
 
-This model specializes in returning a Python List that will give flashcard clues
+More information needed
 
 ## Intended uses & limitations
 
-Its intent is for building flashcards but could(?) be alternatively used to just build python lists of your choice
+More information needed
 
 ## Training and evaluation data
 
-Seen in the academic-flashcards dataset, I used a divy between the two.
+More information needed
 
 ## Training procedure
 
-See the hyperparameters below, I passed those through to transformer's TrainingArguments class which I then used with the Trainer.
-
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
@@ -44,7 +40,7 @@ The following hyperparameters were used during training:
 - eval_batch_size: 16
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: cosine
+- lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
 - num_epochs: 10
 
@@ -52,11 +48,15 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 0.1204        | 1.0   | 1020 | 0.2234          |
-| 0.1091        | 2.0   | 2040 | 0.2287          |
-| 0.0997        | 3.0   | 3060 | 0.2395          |
-
-I had initially forgotten early stopping and had set epochs to 30, unfortunate the overfitting was halted later than I wanted, but could be worse.
+| 0.474         | 1.0   | 1004 | 0.2415          |
+| 0.2877        | 2.0   | 2008 | 0.2146          |
+| 0.2514        | 3.0   | 3012 | 0.2101          |
+| 0.2311        | 4.0   | 4016 | 0.2050          |
+| 0.2131        | 5.0   | 5020 | 0.2004          |
+| 0.2033        | 6.0   | 6024 | 0.2000          |
+| 0.1932        | 7.0   | 7028 | 0.1983          |
+| 0.1849        | 8.0   | 8032 | 0.1997          |
+| 0.1813        | 9.0   | 9036 | 0.1990          |
 
 
 ### Framework versions
@@ -64,4 +64,4 @@ I had initially forgotten early stopping and had set epochs to 30, unfortunate t
 - Transformers 4.51.3
 - Pytorch 2.6.0+cu124
 - Datasets 2.14.4
-- Tokenizers 0.21.1
+- Tokenizers 0.21.1

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f26e65ca69703c6d5d1c1cd3642c3902ba23c31c442b6b8ebcf4a8c04e32fae
+oid sha256:88f2826190116286a60a87a17939303dd06639b49ccc4ef6e7690d65462bb39b
 size 990345064