Update README.md

README.md

@@ -14,7 +14,7 @@ datasets:
 - pfost-bit/surf_forecaster_dataset
 ---
 
-# Model Card for SurfMine
+# SurfMine
 
 
 ## Model Details
@@ -34,7 +34,7 @@ The dataset for this task was created through a pipline that I programmed in pyt
 
 Using the sklearn library I performed a train-test split prior to implementing any training pipeline. I used an 80/10/10 Train/Test/Validation Split. The validation data is unseen by the training at all points until final metrics are calculated.
 
-The dataset can be found [Here](pfost-bit/surf_forecaster_dataset).
+The dataset can be found [Here](https://huggingface.co/datasets/pfost-bit/surf_forecaster_dataset).
 
 ### Methodology
 
@@ -164,4 +164,10 @@ We can see the model response, for the example given above the generated respons
 Hey everyone! There is not much going on behind the shop at the moment. Sets are still breaking close to shore in the knee high range but pretty much nothing bigger than that. Wind is blowing ENE at 13mph keeping some texture on the surface. We are approaching high tide, slotted for 3:06pm. Check back Later!
 ```
 
+## Limitations
 
+The model has a few limitations, notably the input format is pretty clumsy. You need to get input data from NOAA forecasts and feed that as the input into the model, this data, while freely available, is difficult to parse and retrieve. Also, this model is exclusively trained on human forecaster data from one shop in Wrightsville Beach, NC. As a result it is very biased toward conditions in that area. This model will only write about good conditions if they are good for an east facing beach, located in NC. It is not generalized at all. Further, the training data for human forecasters is truncated and ends with [...] every time. This is annoying and I would change the data collection code if I had more time. The purpose of the project was to learn about building a model for a custom task, I have definitely accomplished this!
+
+- Patrick Foster
+- UVASDS
+- ezq9qu@virginia.edu