Upload folder using huggingface_hub

.ipynb_checkpoints/README-checkpoint.md

@@ -0,0 +1,55 @@
+
+# Model Card
+
+## Model Card Authors
+Mathew
+
+## Model Description
+This is a Linear Regression model trained on the UCI Automobile dataset to predict the 'symboling' insurance risk rating from 17 car features including price, horsepower, bore, and curb-weight, amongst other continous variables.
+
+## Intended Uses & Limitations
+This model is for educational purposes only. It is not suitable for production use because the dataset is small (only ~200 or so entries), outdated (~1980s), and contained a lot of missing values (41 missing normalized-losses, around 20% of all rows had a missing normalized-losses entry). Predictions should not be used for real insurance predictions.
+
+## Training Data
+Data source: UCI Automobile dataset (https://archive.ics.uci.edu/dataset/10/automobile). Contains ~200 cars with mixed numeric and categorical features. Missing values were imputed using MICE.
+
+## Evaluation Metrics
+- R2: 0.603
+- RMSE: 0.713
+
+## Ethical Considerations
+The 'symboling' risk value is not only determined by continous, but categorical variables as well, which the model does not account for. While things such as horsepower, bore, engine-size, and number of doors are good predictors, insurance companies also use brands of cars and the type of car (luxury, sport, etc), as well as a variety of other variables to help determine risk factors.
+
+## Audit Questions
+- What features most strongly influence predictions?
+- Are residuals randomly scattered or patterned?
+- How reliable are the evaluation metrics?
+
+
+## Coefficients
+| features          |   coefficients |
+|:------------------|---------------:|
+| price             |   -1.73704e-05 |
+| highway-mpg       |    0.0438076   |
+| city-mpg          |   -0.0610687   |
+| peak-rpm          |   -5.49499e-05 |
+| horsepower        |    0.00207246  |
+| compression-ratio |    0.0187334   |
+| stroke            |   -0.555667    |
+| bore              |   -0.827261    |
+| engine-size       |    0.013724    |
+| num-of-cylinders  |   -0.498651    |
+| curb-weight       |   -5.04019e-05 |
+| height            |    0.0239754   |
+| width             |    0.195005    |
+| length            |    0.0120506   |
+| wheel-base        |   -0.153431    |
+| num-of-doors      |   -0.428882    |
+| normalized-losses |    0.0116676   |
+
+## Plots
+### Predicted vs Actual
+![Predicted vs Actual](predicted_vs_actual.png)
+
+### Residuals Plot
+![Residuals Plot](residuals_plot.png)

README.md

@@ -0,0 +1,55 @@
+
+# Model Card
+
+## Model Card Authors
+Mathew
+
+## Model Description
+This is a Linear Regression model trained on the UCI Automobile dataset to predict the 'symboling' insurance risk rating from 17 car features including price, horsepower, bore, and curb-weight, amongst other continous variables.
+
+## Intended Uses & Limitations
+This model is for educational purposes only. It is not suitable for production use because the dataset is small (only ~200 or so entries), outdated (~1980s), and contained a lot of missing values (41 missing normalized-losses, around 20% of all rows had a missing normalized-losses entry). Predictions should not be used for real insurance predictions.
+
+## Training Data
+Data source: UCI Automobile dataset (https://archive.ics.uci.edu/dataset/10/automobile). Contains ~200 cars with mixed numeric and categorical features. Missing values were imputed using MICE.
+
+## Evaluation Metrics
+- R2: 0.603
+- RMSE: 0.713
+
+## Ethical Considerations
+The 'symboling' risk value is not only determined by continous, but categorical variables as well, which the model does not account for. While things such as horsepower, bore, engine-size, and number of doors are good predictors, insurance companies also use brands of cars and the type of car (luxury, sport, etc), as well as a variety of other variables to help determine risk factors.
+
+## Audit Questions
+- What features most strongly influence predictions?
+- Are residuals randomly scattered or patterned?
+- How reliable are the evaluation metrics?
+
+
+## Coefficients
+| features          |   coefficients |
+|:------------------|---------------:|
+| price             |   -1.73704e-05 |
+| highway-mpg       |    0.0438076   |
+| city-mpg          |   -0.0610687   |
+| peak-rpm          |   -5.49499e-05 |
+| horsepower        |    0.00207246  |
+| compression-ratio |    0.0187334   |
+| stroke            |   -0.555667    |
+| bore              |   -0.827261    |
+| engine-size       |    0.013724    |
+| num-of-cylinders  |   -0.498651    |
+| curb-weight       |   -5.04019e-05 |
+| height            |    0.0239754   |
+| width             |    0.195005    |
+| length            |    0.0120506   |
+| wheel-base        |   -0.153431    |
+| num-of-doors      |   -0.428882    |
+| normalized-losses |    0.0116676   |
+
+## Plots
+### Predicted vs Actual
+![Predicted vs Actual](predicted_vs_actual.png)
+
+### Residuals Plot
+![Residuals Plot](residuals_plot.png)

auto_testing.csv

@@ -0,0 +1,42 @@
+,price,highway-mpg,city-mpg,peak-rpm,horsepower,compression-ratio,stroke,bore,engine-size,num-of-cylinders,curb-weight,height,width,length,wheel-base,num-of-doors,normalized-losses,symboling
+111,15580.0,24,19,5000.0,95.0,8.4,2.19,3.46,120,4,3075,56.7,68.4,186.7,107.9,4.0,161.0,0
+17,36880.0,20,15,5400.0,182.0,8.0,3.39,3.62,209,6,3505,56.3,70.9,197.0,110.0,4.0,128.0,0
+116,17950.0,33,28,4150.0,95.0,21.0,3.52,3.7,152,4,3252,56.7,68.4,186.7,107.9,4.0,161.0,0
+6,17710.0,25,19,5500.0,110.0,8.5,3.4,3.19,136,5,2844,55.7,71.4,192.7,105.8,4.0,158.0,1
+9,17710.0,22,16,5500.0,160.0,7.0,3.4,3.13,131,5,3053,52.0,67.9,178.2,99.5,2.0,197.0,0
+141,7126.0,37,32,4800.0,82.0,9.5,2.64,3.62,108,4,2145,52.5,65.4,172.0,97.2,4.0,102.0,0
+197,16515.0,28,24,5400.0,114.0,9.5,3.15,3.78,141,4,3042,57.5,67.2,188.8,104.3,4.0,74.0,-1
+185,8195.0,34,27,5250.0,85.0,9.0,3.4,3.19,109,4,2212,55.7,65.5,171.7,97.3,4.0,94.0,2
+196,15985.0,28,24,5400.0,114.0,9.5,3.15,3.78,141,4,2935,56.2,67.2,188.8,104.3,4.0,103.0,-2
+127,34028.0,25,17,5900.0,207.0,9.5,2.9,3.74,194,6,2756,51.6,65.0,168.9,89.5,2.0,145.0,3
+183,7975.0,34,27,5250.0,85.0,9.0,3.4,3.19,109,4,2209,55.7,65.5,171.7,97.3,2.0,122.0,2
+62,10245.0,32,26,4800.0,84.0,8.6,3.39,3.39,122,4,2410,55.5,66.5,177.8,98.8,4.0,115.0,0
+187,9495.0,42,37,4500.0,68.0,23.0,3.4,3.01,97,4,2319,55.7,65.5,171.7,97.3,4.0,94.0,2
+20,6575.0,43,38,5400.0,70.0,9.6,3.11,3.03,90,4,1909,52.0,63.6,158.8,94.5,4.0,81.0,0
+135,15510.0,28,21,5250.0,110.0,9.3,3.07,3.54,121,4,2758,56.1,66.5,186.6,99.1,4.0,104.0,2
+162,9258.0,34,28,4800.0,70.0,9.0,3.03,3.19,98,4,2140,52.8,64.4,166.3,95.7,4.0,91.0,0
+45,6338.0,43,38,5400.0,70.0,9.6,3.11,3.03,90,4,1909,52.0,63.6,155.9,94.5,4.0,83.0,0
+83,14869.0,24,19,5000.0,145.0,7.0,3.86,3.59,156,4,2921,50.2,66.3,173.2,95.9,2.0,164.0,3
+129,35056.0,28,17,5750.0,288.0,10.0,3.11,3.94,203,8,3366,50.5,72.3,175.7,98.4,2.0,188.0,1
+2,16500.0,26,19,5000.0,154.0,9.0,3.47,2.68,152,6,2823,52.4,65.5,171.2,94.5,2.0,145.0,1
+40,10295.0,33,27,5800.0,86.0,9.0,3.58,3.15,110,4,2372,54.1,62.5,175.4,96.5,4.0,85.0,0
+52,6795.0,38,31,5000.0,68.0,9.0,3.15,3.03,91,4,1905,54.1,64.2,159.1,93.1,2.0,104.0,1
+75,16503.0,24,19,5000.0,175.0,8.0,3.12,3.78,140,4,2910,54.8,68.0,178.4,102.7,2.0,158.0,1
+13,21105.0,28,21,4250.0,121.0,9.0,3.19,3.31,164,6,2765,54.3,64.8,176.8,101.2,4.0,188.0,0
+171,11549.0,30,24,4800.0,116.0,9.3,3.5,3.62,146,4,2714,52.0,65.6,176.2,98.4,2.0,134.0,2
+21,5572.0,41,37,5500.0,68.0,9.41,3.23,2.97,90,4,1876,50.8,63.8,157.3,93.7,2.0,118.0,1
+54,7395.0,38,31,5000.0,68.0,9.0,3.15,3.08,91,4,1950,54.1,64.2,166.8,93.1,4.0,113.0,1
+42,10345.0,31,25,5500.0,100.0,9.1,3.58,3.15,110,4,2293,51.0,66.0,169.1,96.5,2.0,107.0,1
+194,12940.0,28,23,5400.0,114.0,9.5,3.15,3.78,141,4,2912,56.2,67.2,188.8,104.3,4.0,103.0,-2
+202,21485.0,23,18,5500.0,134.0,8.8,2.87,3.58,173,6,3012,55.5,68.9,188.8,109.1,4.0,95.0,-1
+156,6938.0,37,30,4800.0,70.0,9.0,3.03,3.19,98,4,2081,53.0,64.4,166.3,95.7,4.0,91.0,0
+198,18420.0,22,17,5100.0,162.0,7.5,3.15,3.62,130,4,3045,56.2,67.2,188.8,104.3,4.0,103.0,-2
+150,5348.0,39,35,4800.0,62.0,9.0,3.03,3.05,92,4,1985,54.5,63.6,158.7,95.7,2.0,87.0,1
+147,10198.0,31,25,5200.0,94.0,9.0,2.64,3.62,108,4,2455,53.0,65.4,173.5,97.0,4.0,89.0,0
+19,6295.0,43,38,5400.0,70.0,9.6,3.11,3.03,90,4,1874,52.0,63.6,155.9,94.5,2.0,98.0,1
+108,13200.0,33,28,4150.0,95.0,21.0,3.52,3.7,152,4,3197,56.7,68.4,186.7,107.9,4.0,161.0,0
+168,9639.0,30,24,4800.0,116.0,9.3,3.5,3.62,146,4,2536,52.0,65.6,176.2,98.4,2.0,134.0,2
+22,6377.0,38,31,5500.0,68.0,9.4,3.23,2.97,90,4,1876,50.8,63.8,157.3,93.7,2.0,118.0,1
+140,7603.0,31,26,4400.0,73.0,8.7,2.64,3.62,108,4,2240,55.7,63.8,157.3,93.3,2.0,83.0,2
+199,18950.0,22,17,5100.0,162.0,7.5,3.15,3.62,130,4,3157,57.5,67.2,188.8,104.3,4.0,74.0,-1
+155,8778.0,32,27,4800.0,62.0,9.0,3.03,3.05,92,4,3110,59.1,63.6,169.7,95.7,4.0,91.0,0

config.json

@@ -0,0 +1,117 @@
+{
+    "sklearn": {
+        "columns": [
+            "price",
+            "highway-mpg",
+            "city-mpg",
+            "peak-rpm",
+            "horsepower",
+            "compression-ratio",
+            "stroke",
+            "bore",
+            "engine-size",
+            "num-of-cylinders",
+            "curb-weight",
+            "height",
+            "width",
+            "length",
+            "wheel-base",
+            "num-of-doors",
+            "normalized-losses"
+        ],
+        "environment": [
+            "scikit-learn=1.0.2"
+        ],
+        "example_input": {
+            "price": [
+                13495,
+                16500,
+                13950
+            ],
+            "highway-mpg": [
+                29,
+                28,
+                31
+            ],
+            "city-mpg": [
+                21,
+                19,
+                24
+            ],
+            "peak-rpm": [
+                5000,
+                5500,
+                4800
+            ],
+            "horsepower": [
+                102,
+                115,
+                110
+            ],
+            "compression-ratio": [
+                9.0,
+                9.0,
+                9.0
+            ],
+            "stroke": [
+                3.4,
+                3.4,
+                3.2
+            ],
+            "bore": [
+                3.47,
+                3.01,
+                3.19
+            ],
+            "engine-size": [
+                109,
+                136,
+                120
+            ],
+            "num-of-cylinders": [
+                4,
+                4,
+                4
+            ],
+            "curb-weight": [
+                2548,
+                2823,
+                2507
+            ],
+            "height": [
+                54.3,
+                55.1,
+                54.5
+            ],
+            "width": [
+                64.1,
+                65.5,
+                66.2
+            ],
+            "length": [
+                168.8,
+                171.2,
+                176.6
+            ],
+            "wheel-base": [
+                94.5,
+                94.5,
+                96.5
+            ],
+            "num-of-doors": [
+                4.0,
+                2.0,
+                4.0
+            ],
+            "normalized-losses": [
+                65,
+                103,
+                74
+            ]
+        },
+        "model": {
+            "file": "model.pkl"
+        },
+        "task": "tabular-regression"
+    }
+}

model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:397d94c940c067e31b74c2026c14e70ebed350169a6a9ab44405a86b48a254d6
+size 10632