Bugfix for avg-pool + add some test. (#365)

candle-core/src/cpu_backend.rs

@@ -660,6 +660,8 @@ impl Map1 for AvgPool2D {
                         let mut sum = T::zero();
                         for m in 0..k_h {
                             for n in 0..k_w {
+                                let m = k_h * h_idx + m;
+                                let n = k_w * w_idx + n;
                                 sum += src[src_index + m * stride_h + n * stride_w]
                             }
                         }

candle-core/tests/pool_tests.rs

@@ -0,0 +1,17 @@
+mod test_utils;
+use candle_core::{Device, Tensor};
+
+// https://github.com/huggingface/candle/issues/364
+#[test]
+fn avg_pool2d() -> anyhow::Result<()> {
+    let device = Device::Cpu;
+
+    let data: Vec<f32> = vec![
+        1., 1., 1., 1., 0., 0., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.,
+    ];
+    let t = Tensor::from_vec(data, (1, 1, 4, 4), &device)?;
+
+    let pool = t.avg_pool2d((2, 2), (2, 2))?.squeeze(0)?.squeeze(0)?;
+    assert_eq!(pool.to_vec2::<f32>()?, [[0.5f32, 1.], [1., 1.]]);
+    Ok(())
+}