Update ReadME

JinsooKim-KR · Sep 28, 2023 · d4de284 · d4de284
1 parent 9236c3d
commit d4de284
Showing 1 changed file with 19 additions and 6 deletions.
diff --git a/baselines/FedMeta/README.md b/baselines/FedMeta/README.md
@@ -33,17 +33,30 @@ dataset: [FEMNIST, SHAKESPEARE] # list of datasets you include in your baseline
 * A two-layer CNN network as used in the FedMeta paper (see `models/CNN_Network`). This is the model used by default.
 * A StackedLSTM model used in the FedMeta paper for Shakespeare (see `models/StackedLSTM`).
 
-**You can see more detail model at Apendix.A in paper**
+**You can see more detail at Apendix.A in paper**
 
-****Dataset:**** : This baseline includes the FEMNIST dataset and SHAKESPEARE. Now you should include a breakdown of the details about each of them. Please include information about: how the dataset is partitioned (e.g. LDA with alpha 0.1 as default and all clients have the same number of training examples; or each client gets assigned a different number of samples following a power-law distribution with each client only instances of 2 classes)? if  your dataset is naturally partitioned just state “naturally partitioned”; how many partitions there are (i.e. how many clients)? Please include this an all information relevant about the dataset and its partitioning into a table.
+****Dataset:**** : This baseline includes the FEMNIST dataset and SHAKESPEARE. For data partitioning and sampling per client, we use the Leaf GitHub([LEAF: A Benchmark for Federated Settings](https://github.com/TalwalkarLab/leaf)). The data and client specifications used in this experiment are listed in the table below (Table 1 in the paper). 
 
-|   Dataset   | #Clients | #Samples | #Classes |              partition settings               |
-|:-----------:|:--------:| :---: |:--------:|:---------------------------------------------:|
-|   FEMNIST   |  1,068   | 235,683 |    62    |      Support set : 0.2, Query set : 0.8       |
-| SHAKESPEARE |    110     | 625,127 |    80    |            Support set : 0.2, Query set : 0.8 |
+|   Dataset   | #Clients | #Samples | #Classes |                      #Partition Clients                      | #Partition Dataset                |
+|:-----------:|:--------:| :---: |:--------:|:------------------------------------------------------------:|-----------------------------------|
+|   FEMNIST   |  1,068   | 235,683 |    62    | Train Clients : 0.8, Valid Clients : 0.1, Test Clients : 0.1 | Support set : 0.8, Query set : 0.2 |
+| SHAKESPEARE |    110     | 625,127 |    80    | Train Clients : 0.8, Valid Clients : 0.1, Test Clients : 0.1 | Support set : 0.8, Query set : 0.2|
+
+**The original specifications of the Leaf dataset can be found in the Leaf paper(_"LEAF: A Benchmark for Federated Settings"_).**
 
 ****Training Hyperparameters:**** :warning: *_Include a table with all the main hyperparameters in your baseline. Please show them with their default value._*
 
+|     Algorithm     |    Dataset     |  clients per round  |  number of rounds  | batch size  | optimizer | Learning Rate(α, β) | client resources                     |
+|:-----------------:|:--------------:|:-------------------:|:------------------:|:-----------:|:---------:|:-------------------:|--------------------------------------|
+|      FedAvg       |     FEMNST     | 4(Defalut), 40, 50  |        2000        |     10      |   Adam    |       0.0001        | {'num_cpus': 4.0, 'num_gpus': 0.25 } |
+|      FedAVg       |  SHAKESPEARE   | 4(Defalut), 40, 50  |        400         |     10      |    Adam       |        0.001        | {'num_cpus': 4.0, 'num_gpus': 0.25 } |
+|   FedAvg(Meta)    |     FEMNST     | 4(Defalut), 40, 50  |        2000        |     10      |     Adam      |       0.0001        | {'num_cpus': 4.0, 'num_gpus': 0.25 } |
+|   FedAvg(Meta)    |  SHAKESPEARE   | 4(Defalut), 40, 50  |        400         |     10      |    Adam       |        0.001        | {'num_cpus': 4.0, 'num_gpus': 0.25 } |
+|   FedMeta(MAML)   |     FEMNST     | 4(Defalut), 40, 50  |        2000        |     10      |     Adam      |   (0.001,0.0001)    | {'num_cpus': 4.0, 'num_gpus': 0.25 } |
+|   FedMeta(MAML)   |  SHAKESPEARE   | 4(Defalut), 40, 50  |        400         |     10      |     Adam      |     (0.1,0.01)      | {'num_cpus': 4.0, 'num_gpus': 1.0 }  |
+| FedMeta(Meta-SGD  |     FEMNST     | 4(Defalut), 40, 50  |        2000        |     10      |     Adam      |   (0.001,0.0001)    | {'num_cpus': 4.0, 'num_gpus': 0.25 } |
+| FedMeta(Meta-SGD  |  SHAKESPEARE   | 4(Defalut), 40, 50  |        400         |     10      |    Adam       |     (0.1,0.01)      | {'num_cpus': 4.0, 'num_gpus': 1.0 }  |
+
 
 ## Environment Setup