james-ocf commited on
Commit
2e3ae8e
1 Parent(s): 4fc36a8

Push model using huggingface_hub.

Browse files
Files changed (3) hide show
  1. README.md +52 -0
  2. config.json +31 -0
  3. pytorch_model.bin +3 -0
README.md CHANGED
@@ -1,3 +1,55 @@
1
  ---
 
2
  license: mit
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language: en
3
  license: mit
4
+ library_name: pytorch
5
  ---
6
+
7
+
8
+
9
+
10
+
11
+
12
+ # PVNet2
13
+
14
+ ## Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is/does. -->
17
+ This model class uses satellite data, numericl weather predictions, and recent Grid Service Point( GSP) PV power output to forecast the near-term (~8 hours) PV power output at all GSPs. More information can be found in the model repo [1] and experimental notes in [this google doc](https://docs.google.com/document/d/1fbkfkBzp16WbnCg7RDuRDvgzInA6XQu3xh4NCjV-WDA/edit?usp=sharing).
18
+
19
+ - **Developed by:** openclimatefix
20
+ - **Model type:** Fusion model
21
+ - **Language(s) (NLP):** en
22
+ - **License:** mit
23
+
24
+
25
+ # Training Details
26
+
27
+ ## Data
28
+
29
+ <!-- This should link to a Data 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. -->
30
+
31
+ The model is trained on data from 2017-2020 and validated on data from 2021. See experimental notes in the [the google doc](https://docs.google.com/document/d/1fbkfkBzp16WbnCg7RDuRDvgzInA6XQu3xh4NCjV-WDA/edit?usp=sharing) for more details.
32
+
33
+
34
+ ### Preprocessing
35
+
36
+ Data is prepared with the `ocf_datapipes.training.pvnet` datapipe [2].
37
+
38
+
39
+ ## Results
40
+
41
+ The training logs for the current model can be found [here on wandb](https://wandb.ai/openclimatefix/pvnet2.1/runs/None).
42
+
43
+ The training logs for all model runs of PVNet2 can be found [here](https://wandb.ai/openclimatefix/pvnet2.1).
44
+
45
+ Some experimental notes can be found at in [the google doc](https://docs.google.com/document/d/1fbkfkBzp16WbnCg7RDuRDvgzInA6XQu3xh4NCjV-WDA/edit?usp=sharing)
46
+
47
+
48
+ ### Hardware
49
+
50
+ Trained on a single NVIDIA Tesla T4
51
+
52
+ ### Software
53
+
54
+ - [1] https://github.com/openclimatefix/PVNet
55
+ - [2] https://github.com/openclimatefix/ocf_datapipes
config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_target_": "pvnet_summation.models.model.Model",
3
+ "output_quantiles": [
4
+ 0.1,
5
+ 0.5,
6
+ 0.9
7
+ ],
8
+ "model_name": "openclimatefix/pvnet_v2",
9
+ "model_version": "96ac8c67fa8663844ddcfa82aece51ef94f34453",
10
+ "output_network": {
11
+ "_target_": "pvnet.models.multimodal.linear_networks.networks.ResFCNet2",
12
+ "_partial_": true
13
+ },
14
+ "output_network_kwargs": {
15
+ "fc_hidden_features": 128,
16
+ "n_res_blocks": 2,
17
+ "res_block_layers": 4,
18
+ "dropout_frac": 0.0
19
+ },
20
+ "predict_difference_from_sum": true,
21
+ "optimizer": {
22
+ "_target_": "pvnet.optimizers.AdamWReduceLROnPlateau",
23
+ "lr": 0.0001,
24
+ "weight_decay": 0.5,
25
+ "amsgrad": true,
26
+ "patience": 5,
27
+ "factor": 0.1,
28
+ "threshold": 0.002
29
+ },
30
+ "scale_pvnet_outputs": true
31
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f5d349fcf92021a910b21fd53c8115ce91f3d0c7e510fbad399bd1d01fffba5
3
+ size 65360633