wip modelcard

README.md

@@ -19,12 +19,12 @@ license: mit
 A finetune of Yolo5 intended for identifying beach trash.
 
 - **Developed by:** Jeffrey Queisser, Christopher Buckley
-- **Funded by [optional]:** [More Information Needed]
+- **Funded by [optional]:** Okinawa Institute of Science and Technology COI Next Grant
 - **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
+- **Model type:** Yolo5
 - **Language(s) (NLP):** en
-- **License:** mit
-- **Finetuned from model [optional]:** [More Information Needed]
+- **License:** CC BY-NC 4.0
+- **Finetuned from model [optional]:** Yolo5
 
 ### Model Sources [optional]
 
@@ -80,7 +80,7 @@ Use the code below to get started with the model.
 
 <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
 
-[More Information Needed]
+https://universe.roboflow.com/okinawaaibeachrobot/beach-cleaning-object-detection/dataset/1
 
 ### Training Procedure
 
@@ -88,12 +88,17 @@ Use the code below to get started with the model.
 
 #### Preprocessing [optional]
 
-[More Information Needed]
+
+    {'preprocessing': {'auto-orient': True, 'resize': {'width': 1280, 'height': 800, 'format': 'Fill (with center crop) in'}}}
+    See https://docs.roboflow.com/api-reference/versions/create-a-project-version for more information
+    
 
 
 #### Training Hyperparameters
 
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+- **Training regime:** 
+    TODO populate with any hyperparameters
+     <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
 
 #### Speeds, Sizes, Times [optional]
 
@@ -111,7 +116,7 @@ Use the code below to get started with the model.
 
 <!-- This should link to a Dataset Card if possible. -->
 
-[More Information Needed]
+https://universe.roboflow.com/okinawaaibeachrobot/beach-cleaning-object-detection/dataset/1/images?split=test
 
 #### Factors
 
@@ -123,15 +128,22 @@ Use the code below to get started with the model.
 
 <!-- These are the evaluation metrics being used, ideally with a description of why. -->
 
-[More Information Needed]
+
+    TODO populate with any metrics
+    e.g. loss function
+    
 
 ### Results
 
-[More Information Needed]
+
+    TODO populate with any results
+    
 
 #### Summary
 
 
+    TODO
+    
 
 ## Model Examination [optional]
 
@@ -145,11 +157,11 @@ Use the code below to get started with the model.
 
 Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
 
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
+- **Hardware Type:** TODO
+- **Hours used:** TODO
+- **Cloud Provider:** TODO
+- **Compute Region:** TODO
+- **Carbon Emitted:** TODO
 
 ## Technical Specifications [optional]
 
@@ -193,8 +205,8 @@ Carbon emissions can be estimated using the [Machine Learning Impact calculator]
 
 ## Model Card Authors [optional]
 
-[More Information Needed]
+Christopher Buckley
 
 ## Model Card Contact
 
-[More Information Needed]
+https://github.com/okinawa-ai-beach-robot/home/discussions

modelcard_generate.py

@@ -1,12 +1,47 @@
 from huggingface_hub import ModelCard, ModelCardData
 
-card_data = ModelCardData(language='en', license='mit', library_name='keras')
+card_data = ModelCardData(language="en", license="mit", library_name="keras")
+# See all possible fields at https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md
 card = ModelCard.from_template(
     card_data,
-    model_id='yolo5_beachbot_160',
     model_description="A finetune of Yolo5 intended for identifying beach trash.",
     developers="Jeffrey Queisser, Christopher Buckley",
-    repo="https://github.com/okinawa-ai-beach-robot/yolo5_beachbot_160"
-)
-card.save('README.md')
+    funded_by="Okinawa Institute of Science and Technology COI Next Grant",
+    model_type="Yolo5",
+    license="CC BY-NC 4.0",
+    base_model="Yolo5",
+    repo="https://github.com/okinawa-ai-beach-robot/yolo5_beachbot_160",
+    model_id="yolo5_beachbot_160",
+    training_data="https://universe.roboflow.com/okinawaaibeachrobot/beach-cleaning-object-detection/dataset/1",
+    preprocessing="""
+    {'preprocessing': {'auto-orient': True, 'resize': {'width': 1280, 'height': 800, 'format': 'Fill (with center crop) in'}}}
+    See https://docs.roboflow.com/api-reference/versions/create-a-project-version for more information
+    """,
+    training_regime="""
+    TODO populate with any hyperparameters
+    """,
+    testing_data="https://universe.roboflow.com/okinawaaibeachrobot/beach-cleaning-object-detection/dataset/1/images?split=test",
+    factors="""
+    Note the dataset is small and test set can have quite similar images to the train set. 
+    This was an early vesion of the dataset so results are not reliable",
+    """,
+    testing_metrics="""
+    TODO populate with any metrics
+    e.g. loss function
+    """,
+    results="""
+    TODO populate with any results
+    """,
+    results_summary="""
+    TODO
+    """,
+    hardware_type="TODO",
+    hours_used="TODO",
+    cloud_provider="TODO",
+    cloud_region="TODO",
+    co2_emitted="TODO",
+    model_card_authors="Christopher Buckley",
+    model_card_contact="https://github.com/okinawa-ai-beach-robot/home/discussions"
+    )
+card.save("README.md")
 print(card)